diff options
Diffstat (limited to 'tools/docs')
| -rwxr-xr-x | tools/docs/check-variable-fonts.py | 37 | ||||
| -rwxr-xr-x | tools/docs/checktransupdate.py | 307 | ||||
| -rwxr-xr-x | tools/docs/documentation-file-ref-check | 245 | ||||
| -rwxr-xr-x | tools/docs/features-refresh.sh | 98 | ||||
| -rwxr-xr-x | tools/docs/find-unused-docs.sh | 62 | ||||
| -rwxr-xr-x | tools/docs/get_abi.py | 214 | ||||
| -rwxr-xr-x | tools/docs/get_feat.py | 225 | ||||
| -rw-r--r-- | tools/docs/lib/__init__.py | 0 | ||||
| -rw-r--r-- | tools/docs/lib/enrich_formatter.py | 70 | ||||
| -rwxr-xr-x | tools/docs/lib/parse_data_structs.py | 452 | ||||
| -rwxr-xr-x | tools/docs/list-arch.sh | 11 | ||||
| -rwxr-xr-x | tools/docs/parse-headers.py | 14 | ||||
| -rwxr-xr-x | tools/docs/sphinx-build-wrapper | 864 | ||||
| -rwxr-xr-x | tools/docs/sphinx-pre-install | 1543 | ||||
| -rwxr-xr-x | tools/docs/test_doc_build.py | 513 |
15 files changed, 4126 insertions, 529 deletions
diff --git a/tools/docs/check-variable-fonts.py b/tools/docs/check-variable-fonts.py new file mode 100755 index 000000000000..958d5a745724 --- /dev/null +++ b/tools/docs/check-variable-fonts.py @@ -0,0 +1,37 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0-only +# Copyright (C) Akira Yokosawa, 2024 +# +# Ported to Python by (c) Mauro Carvalho Chehab, 2025 +# +# pylint: disable=C0103 + +""" +Detect problematic Noto CJK variable fonts. + +or more details, see .../tools/lib/python/kdoc/latex_fonts.py. +""" + +import argparse +import sys +import os.path + +src_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, os.path.join(src_dir, '../lib/python')) + +from kdoc.latex_fonts import LatexFontChecker + +checker = LatexFontChecker() + +parser=argparse.ArgumentParser(description=checker.description(), + formatter_class=argparse.RawTextHelpFormatter) +parser.add_argument("--deny-vf", + help="XDG_CONFIG_HOME dir containing fontconfig/fonts.conf file") + +args=parser.parse_args() + +msg = LatexFontChecker(args.deny_vf).check() +if msg: + print(msg) + +sys.exit(1) diff --git a/tools/docs/checktransupdate.py b/tools/docs/checktransupdate.py new file mode 100755 index 000000000000..e894652369a5 --- /dev/null +++ b/tools/docs/checktransupdate.py @@ -0,0 +1,307 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 + +""" +This script helps track the translation status of the documentation +in different locales, e.g., zh_CN. More specially, it uses `git log` +commit to find the latest english commit from the translation commit +(order by author date) and the latest english commits from HEAD. If +differences occur, report the file and commits that need to be updated. + +The usage is as follows: +- tools/docs/checktransupdate.py -l zh_CN +This will print all the files that need to be updated or translated in the zh_CN locale. +- tools/docs/checktransupdate.py Documentation/translations/zh_CN/dev-tools/testing-overview.rst +This will only print the status of the specified file. + +The output is something like: +Documentation/dev-tools/kfence.rst +No translation in the locale of zh_CN + +Documentation/translations/zh_CN/dev-tools/testing-overview.rst +commit 42fb9cfd5b18 ("Documentation: dev-tools: Add link to RV docs") +1 commits needs resolving in total +""" + +import os +import re +import time +import logging +from argparse import ArgumentParser, ArgumentTypeError, BooleanOptionalAction +from datetime import datetime + + +def get_origin_path(file_path): + """Get the origin path from the translation path""" + paths = file_path.split("/") + tidx = paths.index("translations") + opaths = paths[:tidx] + opaths += paths[tidx + 2 :] + return "/".join(opaths) + + +def get_latest_commit_from(file_path, commit): + """Get the latest commit from the specified commit for the specified file""" + command = f"git log --pretty=format:%H%n%aD%n%cD%n%n%B {commit} -1 -- {file_path}" + logging.debug(command) + pipe = os.popen(command) + result = pipe.read() + result = result.split("\n") + if len(result) <= 1: + return None + + logging.debug("Result: %s", result[0]) + + return { + "hash": result[0], + "author_date": datetime.strptime(result[1], "%a, %d %b %Y %H:%M:%S %z"), + "commit_date": datetime.strptime(result[2], "%a, %d %b %Y %H:%M:%S %z"), + "message": result[4:], + } + + +def get_origin_from_trans(origin_path, t_from_head): + """Get the latest origin commit from the translation commit""" + o_from_t = get_latest_commit_from(origin_path, t_from_head["hash"]) + while o_from_t is not None and o_from_t["author_date"] > t_from_head["author_date"]: + o_from_t = get_latest_commit_from(origin_path, o_from_t["hash"] + "^") + if o_from_t is not None: + logging.debug("tracked origin commit id: %s", o_from_t["hash"]) + return o_from_t + + +def get_origin_from_trans_smartly(origin_path, t_from_head): + """Get the latest origin commit from the formatted translation commit: + (1) update to commit HASH (TITLE) + (2) Update the translation through commit HASH (TITLE) + """ + # catch flag for 12-bit commit hash + HASH = r'([0-9a-f]{12})' + # pattern 1: contains "update to commit HASH" + pat_update_to = re.compile(rf'update to commit {HASH}') + # pattern 2: contains "Update the translation through commit HASH" + pat_update_translation = re.compile(rf'Update the translation through commit {HASH}') + + origin_commit_hash = None + for line in t_from_head["message"]: + # check if the line matches the first pattern + match = pat_update_to.search(line) + if match: + origin_commit_hash = match.group(1) + break + # check if the line matches the second pattern + match = pat_update_translation.search(line) + if match: + origin_commit_hash = match.group(1) + break + if origin_commit_hash is None: + return None + o_from_t = get_latest_commit_from(origin_path, origin_commit_hash) + if o_from_t is not None: + logging.debug("tracked origin commit id: %s", o_from_t["hash"]) + return o_from_t + + +def get_commits_count_between(opath, commit1, commit2): + """Get the commits count between two commits for the specified file""" + command = f"git log --pretty=format:%H {commit1}...{commit2} -- {opath}" + logging.debug(command) + pipe = os.popen(command) + result = pipe.read().split("\n") + # filter out empty lines + result = list(filter(lambda x: x != "", result)) + return result + + +def pretty_output(commit): + """Pretty print the commit message""" + command = f"git log --pretty='format:%h (\"%s\")' -1 {commit}" + logging.debug(command) + pipe = os.popen(command) + return pipe.read() + + +def valid_commit(commit): + """Check if the commit is valid or not""" + msg = pretty_output(commit) + return "Merge tag" not in msg + +def check_per_file(file_path): + """Check the translation status for the specified file""" + opath = get_origin_path(file_path) + + if not os.path.isfile(opath): + logging.error("Cannot find the origin path for {file_path}") + return + + o_from_head = get_latest_commit_from(opath, "HEAD") + t_from_head = get_latest_commit_from(file_path, "HEAD") + + if o_from_head is None or t_from_head is None: + logging.error("Cannot find the latest commit for %s", file_path) + return + + o_from_t = get_origin_from_trans_smartly(opath, t_from_head) + # notice, o_from_t from get_*_smartly() is always more accurate than from get_*() + if o_from_t is None: + o_from_t = get_origin_from_trans(opath, t_from_head) + + if o_from_t is None: + logging.error("Error: Cannot find the latest origin commit for %s", file_path) + return + + if o_from_head["hash"] == o_from_t["hash"]: + logging.debug("No update needed for %s", file_path) + else: + logging.info(file_path) + commits = get_commits_count_between( + opath, o_from_t["hash"], o_from_head["hash"] + ) + count = 0 + for commit in commits: + if valid_commit(commit): + logging.info("commit %s", pretty_output(commit)) + count += 1 + logging.info("%d commits needs resolving in total\n", count) + + +def valid_locales(locale): + """Check if the locale is valid or not""" + script_path = os.path.dirname(os.path.abspath(__file__)) + linux_path = os.path.join(script_path, "../..") + if not os.path.isdir(f"{linux_path}/Documentation/translations/{locale}"): + raise ArgumentTypeError("Invalid locale: {locale}") + return locale + + +def list_files_with_excluding_folders(folder, exclude_folders, include_suffix): + """List all files with the specified suffix in the folder and its subfolders""" + files = [] + stack = [folder] + + while stack: + pwd = stack.pop() + # filter out the exclude folders + if os.path.basename(pwd) in exclude_folders: + continue + # list all files and folders + for item in os.listdir(pwd): + ab_item = os.path.join(pwd, item) + if os.path.isdir(ab_item): + stack.append(ab_item) + else: + if ab_item.endswith(include_suffix): + files.append(ab_item) + + return files + + +class DmesgFormatter(logging.Formatter): + """Custom dmesg logging formatter""" + def format(self, record): + timestamp = time.time() + formatted_time = f"[{timestamp:>10.6f}]" + log_message = f"{formatted_time} {record.getMessage()}" + return log_message + + +def config_logging(log_level, log_file="checktransupdate.log"): + """configure logging based on the log level""" + # set up the root logger + logger = logging.getLogger() + logger.setLevel(log_level) + + # Create console handler + console_handler = logging.StreamHandler() + console_handler.setLevel(log_level) + + # Create file handler + file_handler = logging.FileHandler(log_file) + file_handler.setLevel(log_level) + + # Create formatter and add it to the handlers + formatter = DmesgFormatter() + console_handler.setFormatter(formatter) + file_handler.setFormatter(formatter) + + # Add the handler to the logger + logger.addHandler(console_handler) + logger.addHandler(file_handler) + + +def main(): + """Main function of the script""" + script_path = os.path.dirname(os.path.abspath(__file__)) + linux_path = os.path.join(script_path, "../..") + + parser = ArgumentParser(description="Check the translation update") + parser.add_argument( + "-l", + "--locale", + default="zh_CN", + type=valid_locales, + help="Locale to check when files are not specified", + ) + + parser.add_argument( + "--print-missing-translations", + action=BooleanOptionalAction, + default=True, + help="Print files that do not have translations", + ) + + parser.add_argument( + '--log', + default='INFO', + choices=['DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL'], + help='Set the logging level') + + parser.add_argument( + '--logfile', + default='checktransupdate.log', + help='Set the logging file (default: checktransupdate.log)') + + parser.add_argument( + "files", nargs="*", help="Files to check, if not specified, check all files" + ) + args = parser.parse_args() + + # Configure logging based on the --log argument + log_level = getattr(logging, args.log.upper(), logging.INFO) + config_logging(log_level) + + # Get files related to linux path + files = args.files + if len(files) == 0: + offical_files = list_files_with_excluding_folders( + os.path.join(linux_path, "Documentation"), ["translations", "output"], "rst" + ) + + for file in offical_files: + # split the path into parts + path_parts = file.split(os.sep) + # find the index of the "Documentation" directory + kindex = path_parts.index("Documentation") + # insert the translations and locale after the Documentation directory + new_path_parts = path_parts[:kindex + 1] + ["translations", args.locale] \ + + path_parts[kindex + 1 :] + # join the path parts back together + new_file = os.sep.join(new_path_parts) + if os.path.isfile(new_file): + files.append(new_file) + else: + if args.print_missing_translations: + logging.info(os.path.relpath(os.path.abspath(file), linux_path)) + logging.info("No translation in the locale of %s\n", args.locale) + + files = list(map(lambda x: os.path.relpath(os.path.abspath(x), linux_path), files)) + + # cd to linux root directory + os.chdir(linux_path) + + for file in files: + check_per_file(file) + + +if __name__ == "__main__": + main() diff --git a/tools/docs/documentation-file-ref-check b/tools/docs/documentation-file-ref-check new file mode 100755 index 000000000000..0cad42f6943b --- /dev/null +++ b/tools/docs/documentation-file-ref-check @@ -0,0 +1,245 @@ +#!/usr/bin/env perl +# SPDX-License-Identifier: GPL-2.0 +# +# Treewide grep for references to files under Documentation, and report +# non-existing files in stderr. + +use warnings; +use strict; +use Getopt::Long qw(:config no_auto_abbrev); + +# NOTE: only add things here when the file was gone, but the text wants +# to mention a past documentation file, for example, to give credits for +# the original work. +my %false_positives = ( + "Documentation/scsi/scsi_mid_low_api.rst" => "Documentation/Configure.help", + "drivers/vhost/vhost.c" => "Documentation/virtual/lguest/lguest.c", +); + +my $scriptname = $0; +$scriptname =~ s,tools/docs/([^/]+/),$1,; + +# Parse arguments +my $help = 0; +my $fix = 0; +my $warn = 0; + +if (! -e ".git") { + printf "Warning: can't check if file exists, as this is not a git tree\n"; + exit 0; +} + +GetOptions( + 'fix' => \$fix, + 'warn' => \$warn, + 'h|help|usage' => \$help, +); + +if ($help != 0) { + print "$scriptname [--help] [--fix]\n"; + exit -1; +} + +# Step 1: find broken references +print "Finding broken references. This may take a while... " if ($fix); + +my %broken_ref; + +my $doc_fix = 0; + +open IN, "git grep ':doc:\`' Documentation/|" + or die "Failed to run git grep"; +while (<IN>) { + next if (!m,^([^:]+):.*\:doc\:\`([^\`]+)\`,); + next if (m,sphinx/,); + + my $file = $1; + my $d = $1; + my $doc_ref = $2; + + my $f = $doc_ref; + + $d =~ s,(.*/).*,$1,; + $f =~ s,.*\<([^\>]+)\>,$1,; + + if ($f =~ m,^/,) { + $f = "$f.rst"; + $f =~ s,^/,Documentation/,; + } else { + $f = "$d$f.rst"; + } + + next if (grep -e, glob("$f")); + + if ($fix && !$doc_fix) { + print STDERR "\nWARNING: Currently, can't fix broken :doc:`` fields\n"; + } + $doc_fix++; + + print STDERR "$file: :doc:`$doc_ref`\n"; +} +close IN; + +open IN, "git grep 'Documentation/'|" + or die "Failed to run git grep"; +while (<IN>) { + next if (!m/^([^:]+):(.*)/); + + my $f = $1; + my $ln = $2; + + # On linux-next, discard the Next/ directory + next if ($f =~ m,^Next/,); + + # Makefiles and scripts contain nasty expressions to parse docs + next if ($f =~ m/Makefile/ || $f =~ m/\.(sh|py|pl|~|rej|org|orig)$/); + + # It doesn't make sense to parse hidden files + next if ($f =~ m#/\.#); + + # Skip this script + next if ($f eq $scriptname); + + # Ignore the dir where documentation will be built + next if ($ln =~ m,\b(\S*)Documentation/output,); + + if ($ln =~ m,\b(\S*)(Documentation/[A-Za-z0-9\_\.\,\~/\*\[\]\?+-]*)(.*),) { + my $prefix = $1; + my $ref = $2; + my $base = $2; + my $extra = $3; + + # some file references are like: + # /usr/src/linux/Documentation/DMA-{API,mapping}.txt + # For now, ignore them + next if ($extra =~ m/^{/); + + # Remove footnotes at the end like: + # Documentation/devicetree/dt-object-internal.txt[1] + $ref =~ s/(txt|rst)\[\d+]$/$1/; + + # Remove ending ']' without any '[' + $ref =~ s/\].*// if (!($ref =~ m/\[/)); + + # Remove puntuation marks at the end + $ref =~ s/[\,\.]+$//; + + my $fulref = "$prefix$ref"; + + $fulref =~ s/^(\<file|ref)://; + $fulref =~ s/^[\'\`]+//; + $fulref =~ s,^\$\(.*\)/,,; + $base =~ s,.*/,,; + + # Remove URL false-positives + next if ($fulref =~ m/^http/); + + # Remove sched-pelt false-positive + next if ($fulref =~ m,^Documentation/scheduler/sched-pelt$,); + + # Discard some build examples from Documentation/target/tcm_mod_builder.rst + next if ($fulref =~ m,mnt/sdb/lio-core-2.6.git/Documentation/target,); + + # Check if exists, evaluating wildcards + next if (grep -e, glob("$ref $fulref")); + + # Accept relative Documentation patches for tools/ + if ($f =~ m/tools/) { + my $path = $f; + $path =~ s,(.*)/.*,$1,; + $path =~ s,testing/selftests/bpf,bpf/bpftool,; + next if (grep -e, glob("$path/$ref $path/../$ref $path/$fulref")); + } + + # Discard known false-positives + if (defined($false_positives{$f})) { + next if ($false_positives{$f} eq $fulref); + } + + if ($fix) { + if (!($ref =~ m/(scripts|Kconfig|Kbuild)/)) { + $broken_ref{$ref}++; + } + } elsif ($warn) { + print STDERR "Warning: $f references a file that doesn't exist: $fulref\n"; + } else { + print STDERR "$f: $fulref\n"; + } + } +} +close IN; + +exit 0 if (!$fix); + +# Step 2: Seek for file name alternatives +print "Auto-fixing broken references. Please double-check the results\n"; + +foreach my $ref (keys %broken_ref) { + my $new =$ref; + + my $basedir = "."; + # On translations, only seek inside the translations directory + $basedir = $1 if ($ref =~ m,(Documentation/translations/[^/]+),); + + # get just the basename + $new =~ s,.*/,,; + + my $f=""; + + # usual reason for breakage: DT file moved around + if ($ref =~ /devicetree/) { + # usual reason for breakage: DT file renamed to .yaml + if (!$f) { + my $new_ref = $ref; + $new_ref =~ s/\.txt$/.yaml/; + $f=$new_ref if (-f $new_ref); + } + + if (!$f) { + my $search = $new; + $search =~ s,^.*/,,; + $f = qx(find Documentation/devicetree/ -iname "*$search*") if ($search); + if (!$f) { + # Manufacturer name may have changed + $search =~ s/^.*,//; + $f = qx(find Documentation/devicetree/ -iname "*$search*") if ($search); + } + } + } + + # usual reason for breakage: file renamed to .rst + if (!$f) { + $new =~ s/\.txt$/.rst/; + $f=qx(find $basedir -iname $new) if ($new); + } + + # usual reason for breakage: use dash or underline + if (!$f) { + $new =~ s/[-_]/[-_]/g; + $f=qx(find $basedir -iname $new) if ($new); + } + + # Wild guess: seek for the same name on another place + if (!$f) { + $f = qx(find $basedir -iname $new) if ($new); + } + + my @find = split /\s+/, $f; + + if (!$f) { + print STDERR "ERROR: Didn't find a replacement for $ref\n"; + } elsif (scalar(@find) > 1) { + print STDERR "WARNING: Won't auto-replace, as found multiple files close to $ref:\n"; + foreach my $j (@find) { + $j =~ s,^./,,; + print STDERR " $j\n"; + } + } else { + $f = $find[0]; + $f =~ s,^./,,; + print "INFO: Replacing $ref to $f\n"; + foreach my $j (qx(git grep -l $ref)) { + qx(sed "s\@$ref\@$f\@g" -i $j); + } + } +} diff --git a/tools/docs/features-refresh.sh b/tools/docs/features-refresh.sh new file mode 100755 index 000000000000..c2288124e94a --- /dev/null +++ b/tools/docs/features-refresh.sh @@ -0,0 +1,98 @@ +# +# Small script that refreshes the kernel feature support status in place. +# + +for F_FILE in Documentation/features/*/*/arch-support.txt; do + F=$(grep "^# Kconfig:" "$F_FILE" | cut -c26-) + + # + # Each feature F is identified by a pair (O, K), where 'O' can + # be either the empty string (for 'nop') or "not" (the logical + # negation operator '!'); other operators are not supported. + # + O="" + K=$F + if [[ "$F" == !* ]]; then + O="not" + K=$(echo $F | sed -e 's/^!//g') + fi + + # + # F := (O, K) is 'valid' iff there is a Kconfig file (for some + # arch) which contains K. + # + # Notice that this definition entails an 'asymmetry' between + # the case 'O = ""' and the case 'O = "not"'. E.g., F may be + # _invalid_ if: + # + # [case 'O = ""'] + # 1) no arch provides support for F, + # 2) K does not exist (e.g., it was renamed/mis-typed); + # + # [case 'O = "not"'] + # 3) all archs provide support for F, + # 4) as in (2). + # + # The rationale for adopting this definition (and, thus, for + # keeping the asymmetry) is: + # + # We want to be able to 'detect' (2) (or (4)). + # + # (1) and (3) may further warn the developers about the fact + # that K can be removed. + # + F_VALID="false" + for ARCH_DIR in arch/*/; do + K_FILES=$(find $ARCH_DIR -name "Kconfig*") + K_GREP=$(grep "$K" $K_FILES) + if [ ! -z "$K_GREP" ]; then + F_VALID="true" + break + fi + done + if [ "$F_VALID" = "false" ]; then + printf "WARNING: '%s' is not a valid Kconfig\n" "$F" + fi + + T_FILE="$F_FILE.tmp" + grep "^#" $F_FILE > $T_FILE + echo " -----------------------" >> $T_FILE + echo " | arch |status|" >> $T_FILE + echo " -----------------------" >> $T_FILE + for ARCH_DIR in arch/*/; do + ARCH=$(echo $ARCH_DIR | sed -e 's/^arch//g' | sed -e 's/\///g') + K_FILES=$(find $ARCH_DIR -name "Kconfig*") + K_GREP=$(grep "$K" $K_FILES) + # + # Arch support status values for (O, K) are updated according + # to the following rules. + # + # - ("", K) is 'supported by a given arch', if there is a + # Kconfig file for that arch which contains K; + # + # - ("not", K) is 'supported by a given arch', if there is + # no Kconfig file for that arch which contains K; + # + # - otherwise: preserve the previous status value (if any), + # default to 'not yet supported'. + # + # Notice that, according these rules, invalid features may be + # updated/modified. + # + if [ "$O" = "" ] && [ ! -z "$K_GREP" ]; then + printf " |%12s: | ok |\n" "$ARCH" >> $T_FILE + elif [ "$O" = "not" ] && [ -z "$K_GREP" ]; then + printf " |%12s: | ok |\n" "$ARCH" >> $T_FILE + else + S=$(grep -v "^#" "$F_FILE" | grep " $ARCH:") + if [ ! -z "$S" ]; then + echo "$S" >> $T_FILE + else + printf " |%12s: | TODO |\n" "$ARCH" \ + >> $T_FILE + fi + fi + done + echo " -----------------------" >> $T_FILE + mv $T_FILE $F_FILE +done diff --git a/tools/docs/find-unused-docs.sh b/tools/docs/find-unused-docs.sh new file mode 100755 index 000000000000..05552dbda5bc --- /dev/null +++ b/tools/docs/find-unused-docs.sh @@ -0,0 +1,62 @@ +#!/bin/bash +# (c) 2017, Jonathan Corbet <corbet@lwn.net> +# sayli karnik <karniksayli1995@gmail.com> +# +# This script detects files with kernel-doc comments for exported functions +# that are not included in documentation. +# +# usage: Run 'tools/docs/find-unused-docs.sh directory' from top level of kernel +# tree. +# +# example: $tools/docs/find-unused-docs.sh drivers/scsi +# +# Licensed under the terms of the GNU GPL License + +if ! [ -d "Documentation" ]; then + echo "Run from top level of kernel tree" + exit 1 +fi + +if [ "$#" -ne 1 ]; then + echo "Usage: tools/docs/find-unused-docs.sh directory" + exit 1 +fi + +if ! [ -d "$1" ]; then + echo "Directory $1 doesn't exist" + exit 1 +fi + +cd "$( dirname "${BASH_SOURCE[0]}" )" +cd .. + +cd Documentation/ + +echo "The following files contain kerneldoc comments for exported functions \ +that are not used in the formatted documentation" + +# FILES INCLUDED + +files_included=($(grep -rHR ".. kernel-doc" --include \*.rst | cut -d " " -f 3)) + +declare -A FILES_INCLUDED + +for each in "${files_included[@]}"; do + FILES_INCLUDED[$each]="$each" + done + +cd .. + +# FILES NOT INCLUDED + +for file in `find $1 -name '*.c'`; do + + if [[ ${FILES_INCLUDED[$file]+_} ]]; then + continue; + fi + str=$(PYTHONDONTWRITEBYTECODE=1 scripts/kernel-doc -export "$file" 2>/dev/null) + if [[ -n "$str" ]]; then + echo "$file" + fi + done + diff --git a/tools/docs/get_abi.py b/tools/docs/get_abi.py new file mode 100755 index 000000000000..2f0b99401f26 --- /dev/null +++ b/tools/docs/get_abi.py @@ -0,0 +1,214 @@ +#!/usr/bin/env python3 +# pylint: disable=R0903 +# Copyright(c) 2025: Mauro Carvalho Chehab <mchehab@kernel.org>. +# SPDX-License-Identifier: GPL-2.0 + +""" +Parse ABI documentation and produce results from it. +""" + +import argparse +import logging +import os +import sys + +# Import Python modules + +LIB_DIR = "../lib/python" +SRC_DIR = os.path.dirname(os.path.realpath(__file__)) + +sys.path.insert(0, os.path.join(SRC_DIR, LIB_DIR)) + +from abi.abi_parser import AbiParser # pylint: disable=C0413 +from abi.abi_regex import AbiRegex # pylint: disable=C0413 +from abi.helpers import ABI_DIR, DEBUG_HELP # pylint: disable=C0413 +from abi.system_symbols import SystemSymbols # pylint: disable=C0413 + +# Command line classes + + +REST_DESC = """ +Produce output in ReST format. + +The output is done on two sections: + +- Symbols: show all parsed symbols in alphabetic order; +- Files: cross reference the content of each file with the symbols on it. +""" + +class AbiRest: + """Initialize an argparse subparser for rest output""" + + def __init__(self, subparsers): + """Initialize argparse subparsers""" + + parser = subparsers.add_parser("rest", + formatter_class=argparse.RawTextHelpFormatter, + description=REST_DESC) + + parser.add_argument("--enable-lineno", action="store_true", + help="enable lineno") + parser.add_argument("--raw", action="store_true", + help="output text as contained in the ABI files. " + "It not used, output will contain dynamically" + " generated cross references when possible.") + parser.add_argument("--no-file", action="store_true", + help="Don't the files section") + parser.add_argument("--show-hints", help="Show-hints") + + parser.set_defaults(func=self.run) + + def run(self, args): + """Run subparser""" + + parser = AbiParser(args.dir, debug=args.debug) + parser.parse_abi() + parser.check_issues() + + for t in parser.doc(args.raw, not args.no_file): + if args.enable_lineno: + print (f".. LINENO {t[1]}#{t[2]}\n\n") + + print(t[0]) + +class AbiValidate: + """Initialize an argparse subparser for ABI validation""" + + def __init__(self, subparsers): + """Initialize argparse subparsers""" + + parser = subparsers.add_parser("validate", + formatter_class=argparse.ArgumentDefaultsHelpFormatter, + description="list events") + + parser.set_defaults(func=self.run) + + def run(self, args): + """Run subparser""" + + parser = AbiParser(args.dir, debug=args.debug) + parser.parse_abi() + parser.check_issues() + + +class AbiSearch: + """Initialize an argparse subparser for ABI search""" + + def __init__(self, subparsers): + """Initialize argparse subparsers""" + + parser = subparsers.add_parser("search", + formatter_class=argparse.ArgumentDefaultsHelpFormatter, + description="Search ABI using a regular expression") + + parser.add_argument("expression", + help="Case-insensitive search pattern for the ABI symbol") + + parser.set_defaults(func=self.run) + + def run(self, args): + """Run subparser""" + + parser = AbiParser(args.dir, debug=args.debug) + parser.parse_abi() + parser.search_symbols(args.expression) + +UNDEFINED_DESC=""" +Check undefined ABIs on local machine. + +Read sysfs devnodes and check if the devnodes there are defined inside +ABI documentation. + +The search logic tries to minimize the number of regular expressions to +search per each symbol. + +By default, it runs on a single CPU, as Python support for CPU threads +is still experimental, and multi-process runs on Python is very slow. + +On experimental tests, if the number of ABI symbols to search per devnode +is contained on a limit of ~150 regular expressions, using a single CPU +is a lot faster than using multiple processes. However, if the number of +regular expressions to check is at the order of ~30000, using multiple +CPUs speeds up the check. +""" + +class AbiUndefined: + """ + Initialize an argparse subparser for logic to check undefined ABI at + the current machine's sysfs + """ + + def __init__(self, subparsers): + """Initialize argparse subparsers""" + + parser = subparsers.add_parser("undefined", + formatter_class=argparse.RawTextHelpFormatter, + description=UNDEFINED_DESC) + + parser.add_argument("-S", "--sysfs-dir", default="/sys", + help="directory where sysfs is mounted") + parser.add_argument("-s", "--search-string", + help="search string regular expression to limit symbol search") + parser.add_argument("-H", "--show-hints", action="store_true", + help="Hints about definitions for missing ABI symbols.") + parser.add_argument("-j", "--jobs", "--max-workers", type=int, default=1, + help="If bigger than one, enables multiprocessing.") + parser.add_argument("-c", "--max-chunk-size", type=int, default=50, + help="Maximum number of chunk size") + parser.add_argument("-f", "--found", action="store_true", + help="Also show found items. " + "Helpful to debug the parser."), + parser.add_argument("-d", "--dry-run", action="store_true", + help="Don't actually search for undefined. " + "Helpful to debug the parser."), + + parser.set_defaults(func=self.run) + + def run(self, args): + """Run subparser""" + + abi = AbiRegex(args.dir, debug=args.debug, + search_string=args.search_string) + + abi_symbols = SystemSymbols(abi=abi, hints=args.show_hints, + sysfs=args.sysfs_dir) + + abi_symbols.check_undefined_symbols(dry_run=args.dry_run, + found=args.found, + max_workers=args.jobs, + chunk_size=args.max_chunk_size) + + +def main(): + """Main program""" + + parser = argparse.ArgumentParser(formatter_class=argparse.RawTextHelpFormatter) + + parser.add_argument("-d", "--debug", type=int, default=0, help="debug level") + parser.add_argument("-D", "--dir", default=ABI_DIR, help=DEBUG_HELP) + + subparsers = parser.add_subparsers() + + AbiRest(subparsers) + AbiValidate(subparsers) + AbiSearch(subparsers) + AbiUndefined(subparsers) + + args = parser.parse_args() + + if args.debug: + level = logging.DEBUG + else: + level = logging.INFO + + logging.basicConfig(level=level, format="[%(levelname)s] %(message)s") + + if "func" in args: + args.func(args) + else: + sys.exit(f"Please specify a valid command for {sys.argv[0]}") + + +# Call main method +if __name__ == "__main__": + main() diff --git a/tools/docs/get_feat.py b/tools/docs/get_feat.py new file mode 100755 index 000000000000..2b5155a1f134 --- /dev/null +++ b/tools/docs/get_feat.py @@ -0,0 +1,225 @@ +#!/usr/bin/env python3 +# pylint: disable=R0902,R0911,R0912,R0914,R0915 +# Copyright(c) 2025: Mauro Carvalho Chehab <mchehab@kernel.org>. +# SPDX-License-Identifier: GPL-2.0 + + +""" +Parse the Linux Feature files and produce a ReST book. +""" + +import argparse +import os +import subprocess +import sys + +from pprint import pprint + +LIB_DIR = "../../tools/lib/python" +SRC_DIR = os.path.dirname(os.path.realpath(__file__)) + +sys.path.insert(0, os.path.join(SRC_DIR, LIB_DIR)) + +from feat.parse_features import ParseFeature # pylint: disable=C0413 + +SRCTREE = os.path.join(os.path.dirname(os.path.realpath(__file__)), "../..") +DEFAULT_DIR = "Documentation/features" + + +class GetFeature: + """Helper class to parse feature parsing parameters""" + + @staticmethod + def get_current_arch(): + """Detects the current architecture""" + + proc = subprocess.run(["uname", "-m"], check=True, + capture_output=True, text=True) + + arch = proc.stdout.strip() + if arch in ["x86_64", "i386"]: + arch = "x86" + elif arch == "s390x": + arch = "s390" + + return arch + + def run_parser(self, args): + """Execute the feature parser""" + + feat = ParseFeature(args.directory, args.debug, args.enable_fname) + data = feat.parse() + + if args.debug > 2: + pprint(data) + + return feat + + def run_rest(self, args): + """ + Generate tables in ReST format. Three types of tables are + supported, depending on the calling arguments: + + - neither feature nor arch is passed: generates a full matrix; + - arch provided: generates a table of supported tables for the + guiven architecture, eventually filtered by feature; + - only feature provided: generates a table with feature details, + showing what architectures it is implemented. + """ + + feat = self.run_parser(args) + + if args.arch: + rst = feat.output_arch_table(args.arch, args.feat) + elif args.feat: + rst = feat.output_feature(args.feat) + else: + rst = feat.output_matrix() + + print(rst) + + def run_current(self, args): + """ + Instead of using a --arch parameter, get feature for the current + architecture. + """ + + args.arch = self.get_current_arch() + + self.run_rest(args) + + def run_list(self, args): + """ + Generate a list of features for a given architecture, in a format + parseable by other scripts. The output format is not ReST. + """ + + if not args.arch: + args.arch = self.get_current_arch() + + feat = self.run_parser(args) + msg = feat.list_arch_features(args.arch, args.feat) + + print(msg) + + def parse_arch(self, parser): + """Add a --arch parsing argument""" + + parser.add_argument("--arch", + help="Output features for an specific" + " architecture, optionally filtering for a " + "single specific feature.") + + def parse_feat(self, parser): + """Add a --feat parsing argument""" + + parser.add_argument("--feat", "--feature", + help="Output features for a single specific " + "feature.") + + + def current_args(self, subparsers): + """Implementscurrent argparse subparser""" + + parser = subparsers.add_parser("current", + formatter_class=argparse.RawTextHelpFormatter, + description="Output table in ReST " + "compatible ASCII format " + "with features for this " + "machine's architecture") + + self.parse_feat(parser) + parser.set_defaults(func=self.run_current) + + def rest_args(self, subparsers): + """Implement rest argparse subparser""" + + parser = subparsers.add_parser("rest", + formatter_class=argparse.RawTextHelpFormatter, + description="Output table(s) in ReST " + "compatible ASCII format " + "with features in ReST " + "markup language. The " + "output is affected by " + "--arch or --feat/--feature" + " flags.") + + self.parse_arch(parser) + self.parse_feat(parser) + parser.set_defaults(func=self.run_rest) + + def list_args(self, subparsers): + """Implement list argparse subparser""" + + parser = subparsers.add_parser("list", + formatter_class=argparse.RawTextHelpFormatter, + description="List features for this " + "machine's architecture, " + "using an easier to parse " + "format. The output is " + "affected by --arch flag.") + + self.parse_arch(parser) + self.parse_feat(parser) + parser.set_defaults(func=self.run_list) + + def validate_args(self, subparsers): + """Implement validate argparse subparser""" + + parser = subparsers.add_parser("validate", + formatter_class=argparse.RawTextHelpFormatter, + description="Validate the contents of " + "the files under " + f"{DEFAULT_DIR}.") + + parser.set_defaults(func=self.run_parser) + + def parser(self): + """ + Create an arparse with common options and several subparsers + """ + parser = argparse.ArgumentParser(formatter_class=argparse.RawTextHelpFormatter) + + parser.add_argument("-d", "--debug", action="count", default=0, + help="Put the script in verbose mode, useful for " + "debugging. Can be called multiple times, to " + "increase verbosity.") + + parser.add_argument("--directory", "--dir", default=DEFAULT_DIR, + help="Changes the location of the Feature files. " + f"By default, it uses the {DEFAULT_DIR} " + "directory.") + + parser.add_argument("--enable-fname", action="store_true", + help="Prints the file name of the feature files. " + "This can be used in order to track " + "dependencies during documentation build.") + + subparsers = parser.add_subparsers() + + self.current_args(subparsers) + self.rest_args(subparsers) + self.list_args(subparsers) + self.validate_args(subparsers) + + args = parser.parse_args() + + return args + + +def main(): + """Main program""" + + feat = GetFeature() + + args = feat.parser() + + if "func" in args: + args.func(args) + else: + sys.exit(f"Please specify a valid command for {sys.argv[0]}") + + +# Call main method +if __name__ == "__main__": + main() diff --git a/tools/docs/lib/__init__.py b/tools/docs/lib/__init__.py deleted file mode 100644 index e69de29bb2d1..000000000000 --- a/tools/docs/lib/__init__.py +++ /dev/null diff --git a/tools/docs/lib/enrich_formatter.py b/tools/docs/lib/enrich_formatter.py deleted file mode 100644 index bb171567a4ca..000000000000 --- a/tools/docs/lib/enrich_formatter.py +++ /dev/null @@ -1,70 +0,0 @@ -#!/usr/bin/env python3 -# SPDX-License-Identifier: GPL-2.0 -# Copyright (c) 2025 by Mauro Carvalho Chehab <mchehab@kernel.org>. - -""" -Ancillary argparse HelpFormatter class that works on a similar way as -argparse.RawDescriptionHelpFormatter, e.g. description maintains line -breaks, but it also implement transformations to the help text. The -actual transformations ar given by enrich_text(), if the output is tty. - -Currently, the follow transformations are done: - - - Positional arguments are shown in upper cases; - - if output is TTY, ``var`` and positional arguments are shown prepended - by an ANSI SGR code. This is usually translated to bold. On some - terminals, like, konsole, this is translated into a colored bold text. -""" - -import argparse -import re -import sys - -class EnrichFormatter(argparse.HelpFormatter): - """ - Better format the output, making easier to identify the positional args - and how they're used at the __doc__ description. - """ - def __init__(self, *args, **kwargs): - """Initialize class and check if is TTY""" - super().__init__(*args, **kwargs) - self._tty = sys.stdout.isatty() - - def enrich_text(self, text): - """Handle ReST markups (currently, only ``foo``)""" - if self._tty and text: - # Replace ``text`` with ANSI SGR (bold) - return re.sub(r'\`\`(.+?)\`\`', - lambda m: f'\033[1m{m.group(1)}\033[0m', text) - return text - - def _fill_text(self, text, width, indent): - """Enrich descriptions with markups on it""" - enriched = self.enrich_text(text) - return "\n".join(indent + line for line in enriched.splitlines()) - - def _format_usage(self, usage, actions, groups, prefix): - """Enrich positional arguments at usage: line""" - - prog = self._prog - parts = [] - - for action in actions: - if action.option_strings: - opt = action.option_strings[0] - if action.nargs != 0: - opt += f" {action.dest.upper()}" - parts.append(f"[{opt}]") - else: - # Positional argument - parts.append(self.enrich_text(f"``{action.dest.upper()}``")) - - usage_text = f"{prefix or 'usage: '} {prog} {' '.join(parts)}\n" - return usage_text - - def _format_action_invocation(self, action): - """Enrich argument names""" - if not action.option_strings: - return self.enrich_text(f"``{action.dest.upper()}``") - - return ", ".join(action.option_strings) diff --git a/tools/docs/lib/parse_data_structs.py b/tools/docs/lib/parse_data_structs.py deleted file mode 100755 index a5aa2e182052..000000000000 --- a/tools/docs/lib/parse_data_structs.py +++ /dev/null @@ -1,452 +0,0 @@ -#!/usr/bin/env python3 -# SPDX-License-Identifier: GPL-2.0 -# Copyright (c) 2016-2025 by Mauro Carvalho Chehab <mchehab@kernel.org>. -# pylint: disable=R0912,R0915 - -""" -Parse a source file or header, creating ReStructured Text cross references. - -It accepts an optional file to change the default symbol reference or to -suppress symbols from the output. - -It is capable of identifying defines, functions, structs, typedefs, -enums and enum symbols and create cross-references for all of them. -It is also capable of distinguish #define used for specifying a Linux -ioctl. - -The optional rules file contains a set of rules like: - - ignore ioctl VIDIOC_ENUM_FMT - replace ioctl VIDIOC_DQBUF vidioc_qbuf - replace define V4L2_EVENT_MD_FL_HAVE_FRAME_SEQ :c:type:`v4l2_event_motion_det` -""" - -import os -import re -import sys - - -class ParseDataStructs: - """ - Creates an enriched version of a Kernel header file with cross-links - to each C data structure type. - - It is meant to allow having a more comprehensive documentation, where - uAPI headers will create cross-reference links to the code. - - It is capable of identifying defines, functions, structs, typedefs, - enums and enum symbols and create cross-references for all of them. - It is also capable of distinguish #define used for specifying a Linux - ioctl. - - By default, it create rules for all symbols and defines, but it also - allows parsing an exception file. Such file contains a set of rules - using the syntax below: - - 1. Ignore rules: - - ignore <type> <symbol>` - - Removes the symbol from reference generation. - - 2. Replace rules: - - replace <type> <old_symbol> <new_reference> - - Replaces how old_symbol with a new reference. The new_reference can be: - - A simple symbol name; - - A full Sphinx reference. - - On both cases, <type> can be: - - ioctl: for defines that end with _IO*, e.g. ioctl definitions - - define: for other defines - - symbol: for symbols defined within enums; - - typedef: for typedefs; - - enum: for the name of a non-anonymous enum; - - struct: for structs. - - Examples: - - ignore define __LINUX_MEDIA_H - ignore ioctl VIDIOC_ENUM_FMT - replace ioctl VIDIOC_DQBUF vidioc_qbuf - replace define V4L2_EVENT_MD_FL_HAVE_FRAME_SEQ :c:type:`v4l2_event_motion_det` - """ - - # Parser regexes with multiple ways to capture enums and structs - RE_ENUMS = [ - re.compile(r"^\s*enum\s+([\w_]+)\s*\{"), - re.compile(r"^\s*enum\s+([\w_]+)\s*$"), - re.compile(r"^\s*typedef\s*enum\s+([\w_]+)\s*\{"), - re.compile(r"^\s*typedef\s*enum\s+([\w_]+)\s*$"), - ] - RE_STRUCTS = [ - re.compile(r"^\s*struct\s+([_\w][\w\d_]+)\s*\{"), - re.compile(r"^\s*struct\s+([_\w][\w\d_]+)$"), - re.compile(r"^\s*typedef\s*struct\s+([_\w][\w\d_]+)\s*\{"), - re.compile(r"^\s*typedef\s*struct\s+([_\w][\w\d_]+)$"), - ] - - # FIXME: the original code was written a long time before Sphinx C - # domain to have multiple namespaces. To avoid to much turn at the - # existing hyperlinks, the code kept using "c:type" instead of the - # right types. To change that, we need to change the types not only - # here, but also at the uAPI media documentation. - DEF_SYMBOL_TYPES = { - "ioctl": { - "prefix": "\\ ", - "suffix": "\\ ", - "ref_type": ":ref", - "description": "IOCTL Commands", - }, - "define": { - "prefix": "\\ ", - "suffix": "\\ ", - "ref_type": ":ref", - "description": "Macros and Definitions", - }, - # We're calling each definition inside an enum as "symbol" - "symbol": { - "prefix": "\\ ", - "suffix": "\\ ", - "ref_type": ":ref", - "description": "Enumeration values", - }, - "typedef": { - "prefix": "\\ ", - "suffix": "\\ ", - "ref_type": ":c:type", - "description": "Type Definitions", - }, - # This is the description of the enum itself - "enum": { - "prefix": "\\ ", - "suffix": "\\ ", - "ref_type": ":c:type", - "description": "Enumerations", - }, - "struct": { - "prefix": "\\ ", - "suffix": "\\ ", - "ref_type": ":c:type", - "description": "Structures", - }, - } - - def __init__(self, debug: bool = False): - """Initialize internal vars""" - self.debug = debug - self.data = "" - - self.symbols = {} - - for symbol_type in self.DEF_SYMBOL_TYPES: - self.symbols[symbol_type] = {} - - def store_type(self, symbol_type: str, symbol: str, - ref_name: str = None, replace_underscores: bool = True): - """ - Stores a new symbol at self.symbols under symbol_type. - - By default, underscores are replaced by "-" - """ - defs = self.DEF_SYMBOL_TYPES[symbol_type] - - prefix = defs.get("prefix", "") - suffix = defs.get("suffix", "") - ref_type = defs.get("ref_type") - - # Determine ref_link based on symbol type - if ref_type: - if symbol_type == "enum": - ref_link = f"{ref_type}:`{symbol}`" - else: - if not ref_name: - ref_name = symbol.lower() - - # c-type references don't support hash - if ref_type == ":ref" and replace_underscores: - ref_name = ref_name.replace("_", "-") - - ref_link = f"{ref_type}:`{symbol} <{ref_name}>`" - else: - ref_link = symbol - - self.symbols[symbol_type][symbol] = f"{prefix}{ref_link}{suffix}" - - def store_line(self, line): - """Stores a line at self.data, properly indented""" - line = " " + line.expandtabs() - self.data += line.rstrip(" ") - - def parse_file(self, file_in: str): - """Reads a C source file and get identifiers""" - self.data = "" - is_enum = False - is_comment = False - multiline = "" - - with open(file_in, "r", - encoding="utf-8", errors="backslashreplace") as f: - for line_no, line in enumerate(f): - self.store_line(line) - line = line.strip("\n") - - # Handle continuation lines - if line.endswith(r"\\"): - multiline += line[-1] - continue - - if multiline: - line = multiline + line - multiline = "" - - # Handle comments. They can be multilined - if not is_comment: - if re.search(r"/\*.*", line): - is_comment = True - else: - # Strip C99-style comments - line = re.sub(r"(//.*)", "", line) - - if is_comment: - if re.search(r".*\*/", line): - is_comment = False - else: - multiline = line - continue - - # At this point, line variable may be a multilined statement, - # if lines end with \ or if they have multi-line comments - # With that, it can safely remove the entire comments, - # and there's no need to use re.DOTALL for the logic below - - line = re.sub(r"(/\*.*\*/)", "", line) - if not line.strip(): - continue - - # It can be useful for debug purposes to print the file after - # having comments stripped and multi-lines grouped. - if self.debug > 1: - print(f"line {line_no + 1}: {line}") - - # Now the fun begins: parse each type and store it. - - # We opted for a two parsing logic here due to: - # 1. it makes easier to debug issues not-parsed symbols; - # 2. we want symbol replacement at the entire content, not - # just when the symbol is detected. - - if is_enum: - match = re.match(r"^\s*([_\w][\w\d_]+)\s*[\,=]?", line) - if match: - self.store_type("symbol", match.group(1)) - if "}" in line: - is_enum = False - continue - - match = re.match(r"^\s*#\s*define\s+([\w_]+)\s+_IO", line) - if match: - self.store_type("ioctl", match.group(1), - replace_underscores=False) - continue - - match = re.match(r"^\s*#\s*define\s+([\w_]+)(\s+|$)", line) - if match: - self.store_type("define", match.group(1)) - continue - - match = re.match(r"^\s*typedef\s+([_\w][\w\d_]+)\s+(.*)\s+([_\w][\w\d_]+);", - line) - if match: - name = match.group(2).strip() - symbol = match.group(3) - self.store_type("typedef", symbol, ref_name=name) - continue - - for re_enum in self.RE_ENUMS: - match = re_enum.match(line) - if match: - self.store_type("enum", match.group(1)) - is_enum = True - break - - for re_struct in self.RE_STRUCTS: - match = re_struct.match(line) - if match: - self.store_type("struct", match.group(1)) - break - - def process_exceptions(self, fname: str): - """ - Process exceptions file with rules to ignore or replace references. - """ - if not fname: - return - - name = os.path.basename(fname) - - with open(fname, "r", encoding="utf-8", errors="backslashreplace") as f: - for ln, line in enumerate(f): - ln += 1 - line = line.strip() - if not line or line.startswith("#"): - continue - - # Handle ignore rules - match = re.match(r"^ignore\s+(\w+)\s+(\S+)", line) - if match: - c_type = match.group(1) - symbol = match.group(2) - - if c_type not in self.DEF_SYMBOL_TYPES: - sys.exit(f"{name}:{ln}: {c_type} is invalid") - - d = self.symbols[c_type] - if symbol in d: - del d[symbol] - - continue - - # Handle replace rules - match = re.match(r"^replace\s+(\S+)\s+(\S+)\s+(\S+)", line) - if not match: - sys.exit(f"{name}:{ln}: invalid line: {line}") - - c_type, old, new = match.groups() - - if c_type not in self.DEF_SYMBOL_TYPES: - sys.exit(f"{name}:{ln}: {c_type} is invalid") - - reftype = None - - # Parse reference type when the type is specified - - match = re.match(r"^\:c\:(data|func|macro|type)\:\`(.+)\`", new) - if match: - reftype = f":c:{match.group(1)}" - new = match.group(2) - else: - match = re.search(r"(\:ref)\:\`(.+)\`", new) - if match: - reftype = match.group(1) - new = match.group(2) - - # If the replacement rule doesn't have a type, get default - if not reftype: - reftype = self.DEF_SYMBOL_TYPES[c_type].get("ref_type") - if not reftype: - reftype = self.DEF_SYMBOL_TYPES[c_type].get("real_type") - - new_ref = f"{reftype}:`{old} <{new}>`" - - # Change self.symbols to use the replacement rule - if old in self.symbols[c_type]: - self.symbols[c_type][old] = new_ref - else: - print(f"{name}:{ln}: Warning: can't find {old} {c_type}") - - def debug_print(self): - """ - Print debug information containing the replacement rules per symbol. - To make easier to check, group them per type. - """ - if not self.debug: - return - - for c_type, refs in self.symbols.items(): - if not refs: # Skip empty dictionaries - continue - - print(f"{c_type}:") - - for symbol, ref in sorted(refs.items()): - print(f" {symbol} -> {ref}") - - print() - - def gen_output(self): - """Write the formatted output to a file.""" - - # Avoid extra blank lines - text = re.sub(r"\s+$", "", self.data) + "\n" - text = re.sub(r"\n\s+\n", "\n\n", text) - - # Escape Sphinx special characters - text = re.sub(r"([\_\`\*\<\>\&\\\\:\/\|\%\$\#\{\}\~\^])", r"\\\1", text) - - # Source uAPI files may have special notes. Use bold font for them - text = re.sub(r"DEPRECATED", "**DEPRECATED**", text) - - # Delimiters to catch the entire symbol after escaped - start_delim = r"([ \n\t\(=\*\@])" - end_delim = r"(\s|,|\\=|\\:|\;|\)|\}|\{)" - - # Process all reference types - for ref_dict in self.symbols.values(): - for symbol, replacement in ref_dict.items(): - symbol = re.escape(re.sub(r"([\_\`\*\<\>\&\\\\:\/])", r"\\\1", symbol)) - text = re.sub(fr'{start_delim}{symbol}{end_delim}', - fr'\1{replacement}\2', text) - - # Remove "\ " where not needed: before spaces and at the end of lines - text = re.sub(r"\\ ([\n ])", r"\1", text) - text = re.sub(r" \\ ", " ", text) - - return text - - def gen_toc(self): - """ - Create a TOC table pointing to each symbol from the header - """ - text = [] - - # Add header - text.append(".. contents:: Table of Contents") - text.append(" :depth: 2") - text.append(" :local:") - text.append("") - - # Sort symbol types per description - symbol_descriptions = [] - for k, v in self.DEF_SYMBOL_TYPES.items(): - symbol_descriptions.append((v['description'], k)) - - symbol_descriptions.sort() - - # Process each category - for description, c_type in symbol_descriptions: - - refs = self.symbols[c_type] - if not refs: # Skip empty categories - continue - - text.append(f"{description}") - text.append("-" * len(description)) - text.append("") - - # Sort symbols alphabetically - for symbol, ref in sorted(refs.items()): - text.append(f"* :{ref}:") - - text.append("") # Add empty line between categories - - return "\n".join(text) - - def write_output(self, file_in: str, file_out: str, toc: bool): - title = os.path.basename(file_in) - - if toc: - text = self.gen_toc() - else: - text = self.gen_output() - - with open(file_out, "w", encoding="utf-8", errors="backslashreplace") as f: - f.write(".. -*- coding: utf-8; mode: rst -*-\n\n") - f.write(f"{title}\n") - f.write("=" * len(title) + "\n\n") - - if not toc: - f.write(".. parsed-literal::\n\n") - - f.write(text) diff --git a/tools/docs/list-arch.sh b/tools/docs/list-arch.sh new file mode 100755 index 000000000000..96fe83b7058b --- /dev/null +++ b/tools/docs/list-arch.sh @@ -0,0 +1,11 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# Small script that visualizes the kernel feature support status +# of an architecture. +# +# (If no arguments are given then it will print the host architecture's status.) +# + +ARCH=${1:-$(uname -m | sed 's/x86_64/x86/' | sed 's/i386/x86/' | sed 's/s390x/s390/')} + +$(dirname $0)/get_feat.pl list --arch $ARCH diff --git a/tools/docs/parse-headers.py b/tools/docs/parse-headers.py index bfa4e46a53e3..436acea4c6ca 100755 --- a/tools/docs/parse-headers.py +++ b/tools/docs/parse-headers.py @@ -24,10 +24,13 @@ The optional ``FILE_RULES`` contains a set of rules like: replace define V4L2_EVENT_MD_FL_HAVE_FRAME_SEQ :c:type:`v4l2_event_motion_det` """ -import argparse +import argparse, sys +import os.path -from lib.parse_data_structs import ParseDataStructs -from lib.enrich_formatter import EnrichFormatter +src_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, os.path.join(src_dir, '../lib/python')) +from kdoc.parse_data_structs import ParseDataStructs +from kdoc.enrich_formatter import EnrichFormatter def main(): """Main function""" @@ -47,10 +50,7 @@ def main(): args = parser.parse_args() parser = ParseDataStructs(debug=args.debug) - parser.parse_file(args.file_in) - - if args.file_rules: - parser.process_exceptions(args.file_rules) + parser.parse_file(args.file_in, args.file_rules) parser.debug_print() parser.write_output(args.file_in, args.file_out, args.toc) diff --git a/tools/docs/sphinx-build-wrapper b/tools/docs/sphinx-build-wrapper new file mode 100755 index 000000000000..7a5fcef25429 --- /dev/null +++ b/tools/docs/sphinx-build-wrapper @@ -0,0 +1,864 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2025 Mauro Carvalho Chehab <mchehab+huawei@kernel.org> +# +# pylint: disable=R0902, R0912, R0913, R0914, R0915, R0917, C0103 +# +# Converted from docs Makefile and parallel-wrapper.sh, both under +# GPLv2, copyrighted since 2008 by the following authors: +# +# Akira Yokosawa <akiyks@gmail.com> +# Arnd Bergmann <arnd@arndb.de> +# Breno Leitao <leitao@debian.org> +# Carlos Bilbao <carlos.bilbao@amd.com> +# Dave Young <dyoung@redhat.com> +# Donald Hunter <donald.hunter@gmail.com> +# Geert Uytterhoeven <geert+renesas@glider.be> +# Jani Nikula <jani.nikula@intel.com> +# Jan Stancek <jstancek@redhat.com> +# Jonathan Corbet <corbet@lwn.net> +# Joshua Clayton <stillcompiling@gmail.com> +# Kees Cook <keescook@chromium.org> +# Linus Torvalds <torvalds@linux-foundation.org> +# Magnus Damm <damm+renesas@opensource.se> +# Masahiro Yamada <masahiroy@kernel.org> +# Mauro Carvalho Chehab <mchehab+huawei@kernel.org> +# Maxim Cournoyer <maxim.cournoyer@gmail.com> +# Peter Foley <pefoley2@pefoley.com> +# Randy Dunlap <rdunlap@infradead.org> +# Rob Herring <robh@kernel.org> +# Shuah Khan <shuahkh@osg.samsung.com> +# Thorsten Blum <thorsten.blum@toblux.com> +# Tomas Winkler <tomas.winkler@intel.com> + + +""" +Sphinx build wrapper that handles Kernel-specific business rules: + +- it gets the Kernel build environment vars; +- it determines what's the best parallelism; +- it handles SPHINXDIRS + +This tool ensures that MIN_PYTHON_VERSION is satisfied. If version is +below that, it seeks for a new Python version. If found, it re-runs using +the newer version. +""" + +import argparse +import locale +import os +import re +import shlex +import shutil +import subprocess +import sys + +from concurrent import futures +from glob import glob + + +LIB_DIR = "../lib/python" +SRC_DIR = os.path.dirname(os.path.realpath(__file__)) + +sys.path.insert(0, os.path.join(SRC_DIR, LIB_DIR)) + +from kdoc.python_version import PythonVersion +from kdoc.latex_fonts import LatexFontChecker +from jobserver import JobserverExec # pylint: disable=C0413,C0411,E0401 + +# +# Some constants +# +VENV_DEFAULT = "sphinx_latest" +MIN_PYTHON_VERSION = PythonVersion("3.7").version +PAPER = ["", "a4", "letter"] + +TARGETS = { + "cleandocs": { "builder": "clean" }, + "linkcheckdocs": { "builder": "linkcheck" }, + "htmldocs": { "builder": "html" }, + "epubdocs": { "builder": "epub", "out_dir": "epub" }, + "texinfodocs": { "builder": "texinfo", "out_dir": "texinfo" }, + "infodocs": { "builder": "texinfo", "out_dir": "texinfo" }, + "mandocs": { "builder": "man", "out_dir": "man" }, + "latexdocs": { "builder": "latex", "out_dir": "latex" }, + "pdfdocs": { "builder": "latex", "out_dir": "latex" }, + "xmldocs": { "builder": "xml", "out_dir": "xml" }, +} + + +# +# SphinxBuilder class +# + +class SphinxBuilder: + """ + Handles a sphinx-build target, adding needed arguments to build + with the Kernel. + """ + + def get_path(self, path, use_cwd=False, abs_path=False): + """ + Ancillary routine to handle patches the right way, as shell does. + + It first expands "~" and "~user". Then, if patch is not absolute, + join self.srctree. Finally, if requested, convert to abspath. + """ + + path = os.path.expanduser(path) + if not path.startswith("/"): + if use_cwd: + base = os.getcwd() + else: + base = self.srctree + + path = os.path.join(base, path) + + if abs_path: + return os.path.abspath(path) + + return path + + def check_rust(self): + """ + Checks if Rust is enabled + """ + self.rustdoc = False + + config = os.path.join(self.srctree, ".config") + + if not os.path.isfile(config): + return + + re_rust = re.compile(r"CONFIG_RUST=(m|y)") + + try: + with open(config, "r", encoding="utf-8") as fp: + for line in fp: + if re_rust.match(line): + self.rustdoc = True + return + + except OSError as e: + print(f"Failed to open {config}", file=sys.stderr) + + def get_sphinx_extra_opts(self, n_jobs): + """ + Get the number of jobs to be used for docs build passed via command + line and desired sphinx verbosity. + + The number of jobs can be on different places: + + 1) It can be passed via "-j" argument; + 2) The SPHINXOPTS="-j8" env var may have "-j"; + 3) if called via GNU make, -j specifies the desired number of jobs. + with GNU makefile, this number is available via POSIX jobserver; + 4) if none of the above is available, it should default to "-jauto", + and let sphinx decide the best value. + """ + + # + # SPHINXOPTS env var, if used, contains extra arguments to be used + # by sphinx-build time. Among them, it may contain sphinx verbosity + # and desired number of parallel jobs. + # + parser = argparse.ArgumentParser() + parser.add_argument('-j', '--jobs', type=int) + parser.add_argument('-q', '--quiet', action='store_true') + + # + # Other sphinx-build arguments go as-is, so place them + # at self.sphinxopts, using shell parser + # + sphinxopts = shlex.split(os.environ.get("SPHINXOPTS", "")) + + # + # Build a list of sphinx args, honoring verbosity here if specified + # + + verbose = self.verbose + sphinx_args, self.sphinxopts = parser.parse_known_args(sphinxopts) + if sphinx_args.quiet is True: + verbose = False + + # + # If the user explicitly sets "-j" at command line, use it. + # Otherwise, pick it from SPHINXOPTS args + # + if n_jobs: + self.n_jobs = n_jobs + elif sphinx_args.jobs: + self.n_jobs = sphinx_args.jobs + else: + self.n_jobs = None + + if not verbose: + self.sphinxopts += ["-q"] + + def __init__(self, builddir, venv=None, verbose=False, n_jobs=None, + interactive=None): + """Initialize internal variables""" + self.venv = venv + self.verbose = None + + # + # Normal variables passed from Kernel's makefile + # + self.kernelversion = os.environ.get("KERNELVERSION", "unknown") + self.kernelrelease = os.environ.get("KERNELRELEASE", "unknown") + self.pdflatex = os.environ.get("PDFLATEX", "xelatex") + + # + # Kernel main Makefile defines a PYTHON3 variable whose default is + # "python3". When set to a different value, it allows running a + # diferent version than the default official python3 package. + # Several distros package python3xx-sphinx packages with newer + # versions of Python and sphinx-build. + # + # Honor such variable different than default + # + self.python = os.environ.get("PYTHON3") + if self.python == "python3": + self.python = None + + if not interactive: + self.latexopts = os.environ.get("LATEXOPTS", "-interaction=batchmode -no-shell-escape") + else: + self.latexopts = os.environ.get("LATEXOPTS", "") + + if not verbose: + verbose = bool(os.environ.get("KBUILD_VERBOSE", "") != "") + + if verbose is not None: + self.verbose = verbose + + # + # Source tree directory. This needs to be at os.environ, as + # Sphinx extensions use it + # + self.srctree = os.environ.get("srctree") + if not self.srctree: + self.srctree = "." + os.environ["srctree"] = self.srctree + + # + # Now that we can expand srctree, get other directories as well + # + self.sphinxbuild = os.environ.get("SPHINXBUILD", "sphinx-build") + self.kerneldoc = self.get_path(os.environ.get("KERNELDOC", + "scripts/kernel-doc.py")) + self.builddir = self.get_path(builddir, use_cwd=True, abs_path=True) + + # + # Get directory locations for LaTeX build toolchain + # + self.pdflatex_cmd = shutil.which(self.pdflatex) + self.latexmk_cmd = shutil.which("latexmk") + + self.env = os.environ.copy() + + self.get_sphinx_extra_opts(n_jobs) + + self.check_rust() + + # + # If venv command line argument is specified, run Sphinx from venv + # + if venv: + bin_dir = os.path.join(venv, "bin") + if not os.path.isfile(os.path.join(bin_dir, "activate")): + sys.exit(f"Venv {venv} not found.") + + # "activate" virtual env + self.env["PATH"] = bin_dir + ":" + self.env["PATH"] + self.env["VIRTUAL_ENV"] = venv + if "PYTHONHOME" in self.env: + del self.env["PYTHONHOME"] + print(f"Setting venv to {venv}") + + def run_sphinx(self, sphinx_build, build_args, *args, **pwargs): + """ + Executes sphinx-build using current python3 command. + + When calling via GNU make, POSIX jobserver is used to tell how + many jobs are still available from a job pool. claim all remaining + jobs, as we don't want sphinx-build to run in parallel with other + jobs. + + Despite that, the user may actually force a different value than + the number of available jobs via command line. + + The "with" logic here is used to ensure that the claimed jobs will + be freed once subprocess finishes + """ + + with JobserverExec() as jobserver: + if jobserver.claim: + # + # when GNU make is used, claim available jobs from jobserver + # + n_jobs = str(jobserver.claim) + else: + # + # Otherwise, let sphinx decide by default + # + n_jobs = "auto" + + # + # If explicitly requested via command line, override default + # + if self.n_jobs: + n_jobs = str(self.n_jobs) + + # + # We can't simply call python3 sphinx-build, as OpenSUSE + # Tumbleweed uses an ELF binary file (/usr/bin/alts) to switch + # between different versions of sphinx-build. So, only call it + # prepending "python3.xx" when PYTHON3 variable is not default. + # + if self.python: + cmd = [self.python] + else: + cmd = [] + + cmd += [sphinx_build] + cmd += [f"-j{n_jobs}"] + cmd += build_args + cmd += self.sphinxopts + + if self.verbose: + print(" ".join(cmd)) + + return subprocess.call(cmd, *args, **pwargs) + + def handle_html(self, css, output_dir): + """ + Extra steps for HTML and epub output. + + For such targets, we need to ensure that CSS will be properly + copied to the output _static directory + """ + + if css: + css = os.path.expanduser(css) + if not css.startswith("/"): + css = os.path.join(self.srctree, css) + + static_dir = os.path.join(output_dir, "_static") + os.makedirs(static_dir, exist_ok=True) + + try: + shutil.copy2(css, static_dir) + except (OSError, IOError) as e: + print(f"Warning: Failed to copy CSS: {e}", file=sys.stderr) + + if self.rustdoc: + print("Building rust docs") + if "MAKE" in self.env: + cmd = [self.env["MAKE"]] + else: + cmd = ["make", "LLVM=1"] + + cmd += [ "rustdoc"] + if self.verbose: + print(" ".join(cmd)) + + try: + subprocess.run(cmd, check=True) + except subprocess.CalledProcessError as e: + print(f"Ignored errors when building rustdoc: {e}. Is RUST enabled?", + file=sys.stderr) + + def build_pdf_file(self, latex_cmd, from_dir, path): + """Builds a single pdf file using latex_cmd""" + try: + subprocess.run(latex_cmd + [path], + cwd=from_dir, check=True, env=self.env) + + return True + except subprocess.CalledProcessError: + return False + + def pdf_parallel_build(self, tex_suffix, latex_cmd, tex_files, n_jobs): + """Build PDF files in parallel if possible""" + builds = {} + build_failed = False + max_len = 0 + has_tex = False + + # + # LaTeX PDF error code is almost useless for us: + # any warning makes it non-zero. For kernel doc builds it always return + # non-zero even when build succeeds. So, let's do the best next thing: + # Ignore build errors. At the end, check if all PDF files were built, + # printing a summary with the built ones and returning 0 if all of + # them were actually built. + # + with futures.ThreadPoolExecutor(max_workers=n_jobs) as executor: + jobs = {} + + for from_dir, pdf_dir, entry in tex_files: + name = entry.name + + if not name.endswith(tex_suffix): + continue + + name = name[:-len(tex_suffix)] + has_tex = True + + future = executor.submit(self.build_pdf_file, latex_cmd, + from_dir, entry.path) + jobs[future] = (from_dir, pdf_dir, name) + + for future in futures.as_completed(jobs): + from_dir, pdf_dir, name = jobs[future] + + pdf_name = name + ".pdf" + pdf_from = os.path.join(from_dir, pdf_name) + pdf_to = os.path.join(pdf_dir, pdf_name) + out_name = os.path.relpath(pdf_to, self.builddir) + max_len = max(max_len, len(out_name)) + + try: + success = future.result() + + if success and os.path.exists(pdf_from): + os.rename(pdf_from, pdf_to) + + # + # if verbose, get the name of built PDF file + # + if self.verbose: + builds[out_name] = "SUCCESS" + else: + builds[out_name] = "FAILED" + build_failed = True + except futures.Error as e: + builds[out_name] = f"FAILED ({repr(e)})" + build_failed = True + + # + # Handle case where no .tex files were found + # + if not has_tex: + out_name = "LaTeX files" + max_len = max(max_len, len(out_name)) + builds[out_name] = "FAILED: no .tex files were generated" + build_failed = True + + return builds, build_failed, max_len + + def handle_pdf(self, output_dirs, deny_vf): + """ + Extra steps for PDF output. + + As PDF is handled via a LaTeX output, after building the .tex file, + a new build is needed to create the PDF output from the latex + directory. + """ + builds = {} + max_len = 0 + tex_suffix = ".tex" + tex_files = [] + + # + # Since early 2024, Fedora and openSUSE tumbleweed have started + # deploying variable-font format of "Noto CJK", causing LaTeX + # to break with CJK. Work around it, by denying the variable font + # usage during xelatex build by passing the location of a config + # file with a deny list. + # + # See tools/docs/lib/latex_fonts.py for more details. + # + if deny_vf: + deny_vf = os.path.expanduser(deny_vf) + if os.path.isdir(deny_vf): + self.env["XDG_CONFIG_HOME"] = deny_vf + + for from_dir in output_dirs: + pdf_dir = os.path.join(from_dir, "../pdf") + os.makedirs(pdf_dir, exist_ok=True) + + if self.latexmk_cmd: + latex_cmd = [self.latexmk_cmd, f"-{self.pdflatex}"] + else: + latex_cmd = [self.pdflatex] + + latex_cmd.extend(shlex.split(self.latexopts)) + + # Get a list of tex files to process + with os.scandir(from_dir) as it: + for entry in it: + if entry.name.endswith(tex_suffix): + tex_files.append((from_dir, pdf_dir, entry)) + + # + # When using make, this won't be used, as the number of jobs comes + # from POSIX jobserver. So, this covers the case where build comes + # from command line. On such case, serialize by default, except if + # the user explicitly sets the number of jobs. + # + n_jobs = 1 + + # n_jobs is either an integer or "auto". Only use it if it is a number + if self.n_jobs: + try: + n_jobs = int(self.n_jobs) + except ValueError: + pass + + # + # When using make, jobserver.claim is the number of jobs that were + # used with "-j" and that aren't used by other make targets + # + with JobserverExec() as jobserver: + n_jobs = 1 + + # + # Handle the case when a parameter is passed via command line, + # using it as default, if jobserver doesn't claim anything + # + if self.n_jobs: + try: + n_jobs = int(self.n_jobs) + except ValueError: + pass + + if jobserver.claim: + n_jobs = jobserver.claim + + builds, build_failed, max_len = self.pdf_parallel_build(tex_suffix, + latex_cmd, + tex_files, + n_jobs) + + # + # In verbose mode, print a summary with the build results per file. + # Otherwise, print a single line with all failures, if any. + # On both cases, return code 1 indicates build failures, + # + if self.verbose: + msg = "Summary" + msg += "\n" + "=" * len(msg) + print() + print(msg) + + for pdf_name, pdf_file in builds.items(): + print(f"{pdf_name:<{max_len}}: {pdf_file}") + + print() + if build_failed: + msg = LatexFontChecker().check() + if msg: + print(msg) + + sys.exit("Error: not all PDF files were created.") + + elif build_failed: + n_failures = len(builds) + failures = ", ".join(builds.keys()) + + msg = LatexFontChecker().check() + if msg: + print(msg) + + sys.exit(f"Error: Can't build {n_failures} PDF file(s): {failures}") + + def handle_info(self, output_dirs): + """ + Extra steps for Info output. + + For texinfo generation, an additional make is needed from the + texinfo directory. + """ + + for output_dir in output_dirs: + try: + subprocess.run(["make", "info"], cwd=output_dir, check=True) + except subprocess.CalledProcessError as e: + sys.exit(f"Error generating info docs: {e}") + + def handle_man(self, kerneldoc, docs_dir, src_dir, output_dir): + """ + Create man pages from kernel-doc output + """ + + re_kernel_doc = re.compile(r"^\.\.\s+kernel-doc::\s*(\S+)") + re_man = re.compile(r'^\.TH "[^"]*" (\d+) "([^"]*)"') + + if docs_dir == src_dir: + # + # Pick the entire set of kernel-doc markups from the entire tree + # + kdoc_files = set([self.srctree]) + else: + kdoc_files = set() + + for fname in glob(os.path.join(src_dir, "**"), recursive=True): + if os.path.isfile(fname) and fname.endswith(".rst"): + with open(fname, "r", encoding="utf-8") as in_fp: + data = in_fp.read() + + for line in data.split("\n"): + match = re_kernel_doc.match(line) + if match: + if os.path.isfile(match.group(1)): + kdoc_files.add(match.group(1)) + + if not kdoc_files: + sys.exit(f"Directory {src_dir} doesn't contain kernel-doc tags") + + cmd = [ kerneldoc, "-m" ] + sorted(kdoc_files) + try: + if self.verbose: + print(" ".join(cmd)) + + result = subprocess.run(cmd, stdout=subprocess.PIPE, text= True) + + if result.returncode: + print(f"Warning: kernel-doc returned {result.returncode} warnings") + + except (OSError, ValueError, subprocess.SubprocessError) as e: + sys.exit(f"Failed to create man pages for {src_dir}: {repr(e)}") + + fp = None + try: + for line in result.stdout.split("\n"): + match = re_man.match(line) + if not match: + if fp: + fp.write(line + '\n') + continue + + if fp: + fp.close() + + fname = f"{output_dir}/{match.group(2)}.{match.group(1)}" + + if self.verbose: + print(f"Creating {fname}") + fp = open(fname, "w", encoding="utf-8") + fp.write(line + '\n') + finally: + if fp: + fp.close() + + def cleandocs(self, builder): # pylint: disable=W0613 + """Remove documentation output directory""" + shutil.rmtree(self.builddir, ignore_errors=True) + + def build(self, target, sphinxdirs=None, + theme=None, css=None, paper=None, deny_vf=None, + skip_sphinx=False): + """ + Build documentation using Sphinx. This is the core function of this + module. It prepares all arguments required by sphinx-build. + """ + + builder = TARGETS[target]["builder"] + out_dir = TARGETS[target].get("out_dir", "") + + # + # Cleandocs doesn't require sphinx-build + # + if target == "cleandocs": + self.cleandocs(builder) + return + + if theme: + os.environ["DOCS_THEME"] = theme + + # + # Other targets require sphinx-build, so check if it exists + # + if not skip_sphinx: + sphinxbuild = shutil.which(self.sphinxbuild, path=self.env["PATH"]) + if not sphinxbuild and target != "mandocs": + sys.exit(f"Error: {self.sphinxbuild} not found in PATH.\n") + + if target == "pdfdocs": + if not self.pdflatex_cmd and not self.latexmk_cmd: + sys.exit("Error: pdflatex or latexmk required for PDF generation") + + docs_dir = os.path.abspath(os.path.join(self.srctree, "Documentation")) + + # + # Fill in base arguments for Sphinx build + # + kerneldoc = self.kerneldoc + if kerneldoc.startswith(self.srctree): + kerneldoc = os.path.relpath(kerneldoc, self.srctree) + + args = [ "-b", builder, "-c", docs_dir ] + + if builder == "latex": + if not paper: + paper = PAPER[1] + + args.extend(["-D", f"latex_elements.papersize={paper}paper"]) + + if self.rustdoc: + args.extend(["-t", "rustdoc"]) + + if not sphinxdirs: + sphinxdirs = os.environ.get("SPHINXDIRS", ".") + + # + # The sphinx-build tool has a bug: internally, it tries to set + # locale with locale.setlocale(locale.LC_ALL, ''). This causes a + # crash if language is not set. Detect and fix it. + # + try: + locale.setlocale(locale.LC_ALL, '') + except locale.Error: + self.env["LC_ALL"] = "C" + + # + # sphinxdirs can be a list or a whitespace-separated string + # + sphinxdirs_list = [] + for sphinxdir in sphinxdirs: + if isinstance(sphinxdir, list): + sphinxdirs_list += sphinxdir + else: + sphinxdirs_list += sphinxdir.split() + + # + # Step 1: Build each directory in separate. + # + # This is not the best way of handling it, as cross-references between + # them will be broken, but this is what we've been doing since + # the beginning. + # + output_dirs = [] + for sphinxdir in sphinxdirs_list: + src_dir = os.path.join(docs_dir, sphinxdir) + doctree_dir = os.path.join(self.builddir, ".doctrees") + output_dir = os.path.join(self.builddir, sphinxdir, out_dir) + + # + # Make directory names canonical + # + src_dir = os.path.normpath(src_dir) + doctree_dir = os.path.normpath(doctree_dir) + output_dir = os.path.normpath(output_dir) + + os.makedirs(doctree_dir, exist_ok=True) + os.makedirs(output_dir, exist_ok=True) + + output_dirs.append(output_dir) + + build_args = args + [ + "-d", doctree_dir, + "-D", f"kerneldoc_bin={kerneldoc}", + "-D", f"version={self.kernelversion}", + "-D", f"release={self.kernelrelease}", + "-D", f"kerneldoc_srctree={self.srctree}", + src_dir, + output_dir, + ] + + if target == "mandocs": + self.handle_man(kerneldoc, docs_dir, src_dir, output_dir) + elif not skip_sphinx: + try: + result = self.run_sphinx(sphinxbuild, build_args, + env=self.env) + + if result: + sys.exit(f"Build failed: return code: {result}") + + except (OSError, ValueError, subprocess.SubprocessError) as e: + sys.exit(f"Build failed: {repr(e)}") + + # + # Ensure that each html/epub output will have needed static files + # + if target in ["htmldocs", "epubdocs"]: + self.handle_html(css, output_dir) + + # + # Step 2: Some targets (PDF and info) require an extra step once + # sphinx-build finishes + # + if target == "pdfdocs": + self.handle_pdf(output_dirs, deny_vf) + elif target == "infodocs": + self.handle_info(output_dirs) + +def jobs_type(value): + """ + Handle valid values for -j. Accepts Sphinx "-jauto", plus a number + equal or bigger than one. + """ + if value is None: + return None + + if value.lower() == 'auto': + return value.lower() + + try: + if int(value) >= 1: + return value + + raise argparse.ArgumentTypeError(f"Minimum jobs is 1, got {value}") + except ValueError: + raise argparse.ArgumentTypeError(f"Must be 'auto' or positive integer, got {value}") # pylint: disable=W0707 + +def main(): + """ + Main function. The only mandatory argument is the target. If not + specified, the other arguments will use default values if not + specified at os.environ. + """ + parser = argparse.ArgumentParser(description="Kernel documentation builder") + + parser.add_argument("target", choices=list(TARGETS.keys()), + help="Documentation target to build") + parser.add_argument("--sphinxdirs", nargs="+", + help="Specific directories to build") + parser.add_argument("--builddir", default="output", + help="Sphinx configuration file") + + parser.add_argument("--theme", help="Sphinx theme to use") + + parser.add_argument("--css", help="Custom CSS file for HTML/EPUB") + + parser.add_argument("--paper", choices=PAPER, default=PAPER[0], + help="Paper size for LaTeX/PDF output") + + parser.add_argument('--deny-vf', + help="Configuration to deny variable fonts on pdf builds") + + parser.add_argument("-v", "--verbose", action='store_true', + help="place build in verbose mode") + + parser.add_argument('-j', '--jobs', type=jobs_type, + help="Sets number of jobs to use with sphinx-build") + + parser.add_argument('-i', '--interactive', action='store_true', + help="Change latex default to run in interactive mode") + + parser.add_argument('-s', '--skip-sphinx-build', action='store_true', + help="Skip sphinx-build step") + + parser.add_argument("-V", "--venv", nargs='?', const=f'{VENV_DEFAULT}', + default=None, + help=f'If used, run Sphinx from a venv dir (default dir: {VENV_DEFAULT})') + + args = parser.parse_args() + + PythonVersion.check_python(MIN_PYTHON_VERSION, show_alternatives=True, + bail_out=True) + + builder = SphinxBuilder(builddir=args.builddir, venv=args.venv, + verbose=args.verbose, n_jobs=args.jobs, + interactive=args.interactive) + + builder.build(args.target, sphinxdirs=args.sphinxdirs, + theme=args.theme, css=args.css, paper=args.paper, + deny_vf=args.deny_vf, + skip_sphinx=args.skip_sphinx_build) + +if __name__ == "__main__": + main() diff --git a/tools/docs/sphinx-pre-install b/tools/docs/sphinx-pre-install new file mode 100755 index 000000000000..965c9b093a41 --- /dev/null +++ b/tools/docs/sphinx-pre-install @@ -0,0 +1,1543 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0-or-later +# Copyright (c) 2017-2025 Mauro Carvalho Chehab <mchehab+huawei@kernel.org> +# +# pylint: disable=C0103,C0114,C0115,C0116,C0301,C0302 +# pylint: disable=R0902,R0904,R0911,R0912,R0914,R0915,R1705,R1710,E1121 + +# Note: this script requires at least Python 3.6 to run. +# Don't add changes not compatible with it, it is meant to report +# incompatible python versions. + +""" +Dependency checker for Sphinx documentation Kernel build. + +This module provides tools to check for all required dependencies needed to +build documentation using Sphinx, including system packages, Python modules +and LaTeX packages for PDF generation. + +It detect packages for a subset of Linux distributions used by Kernel +maintainers, showing hints and missing dependencies. + +The main class SphinxDependencyChecker handles the dependency checking logic +and provides recommendations for installing missing packages. It supports both +system package installations and Python virtual environments. By default, +system pacage install is recommended. +""" + +import argparse +import locale +import os +import re +import subprocess +import sys +from glob import glob +import os.path + +src_dir = os.path.dirname(os.path.realpath(__file__)) +sys.path.insert(0, os.path.join(src_dir, '../lib/python')) +from kdoc.python_version import PythonVersion + +RECOMMENDED_VERSION = PythonVersion("3.4.3").version +MIN_PYTHON_VERSION = PythonVersion("3.7").version + + +class DepManager: + """ + Manage package dependencies. There are three types of dependencies: + + - System: dependencies required for docs build; + - Python: python dependencies for a native distro Sphinx install; + - PDF: dependencies needed by PDF builds. + + Each dependency can be mandatory or optional. Not installing an optional + dependency won't break the build, but will cause degradation at the + docs output. + """ + + # Internal types of dependencies. Don't use them outside DepManager class. + _SYS_TYPE = 0 + _PHY_TYPE = 1 + _PDF_TYPE = 2 + + # Dependencies visible outside the class. + # The keys are tuple with: (type, is_mandatory flag). + # + # Currently we're not using all optional dep types. Yet, we'll keep all + # possible combinations here. They're not many, and that makes easier + # if later needed and for the name() method below + + SYSTEM_MANDATORY = (_SYS_TYPE, True) + PYTHON_MANDATORY = (_PHY_TYPE, True) + PDF_MANDATORY = (_PDF_TYPE, True) + + SYSTEM_OPTIONAL = (_SYS_TYPE, False) + PYTHON_OPTIONAL = (_PHY_TYPE, False) + PDF_OPTIONAL = (_PDF_TYPE, True) + + def __init__(self, pdf): + """ + Initialize internal vars: + + - missing: missing dependencies list, containing a distro-independent + name for a missing dependency and its type. + - missing_pkg: ancillary dict containing missing dependencies in + distro namespace, organized by type. + - need: total number of needed dependencies. Never cleaned. + - optional: total number of optional dependencies. Never cleaned. + - pdf: Is PDF support enabled? + """ + self.missing = {} + self.missing_pkg = {} + self.need = 0 + self.optional = 0 + self.pdf = pdf + + @staticmethod + def name(dtype): + """ + Ancillary routine to output a warn/error message reporting + missing dependencies. + """ + if dtype[0] == DepManager._SYS_TYPE: + msg = "build" + elif dtype[0] == DepManager._PHY_TYPE: + msg = "Python" + else: + msg = "PDF" + + if dtype[1]: + return f"ERROR: {msg} mandatory deps missing" + else: + return f"Warning: {msg} optional deps missing" + + @staticmethod + def is_optional(dtype): + """Ancillary routine to report if a dependency is optional""" + return not dtype[1] + + @staticmethod + def is_pdf(dtype): + """Ancillary routine to report if a dependency is for PDF generation""" + if dtype[0] == DepManager._PDF_TYPE: + return True + + return False + + def add_package(self, package, dtype): + """ + Add a package at the self.missing() dictionary. + Doesn't update missing_pkg. + """ + is_optional = DepManager.is_optional(dtype) + self.missing[package] = dtype + if is_optional: + self.optional += 1 + else: + self.need += 1 + + def del_package(self, package): + """ + Remove a package at the self.missing() dictionary. + Doesn't update missing_pkg. + """ + if package in self.missing: + del self.missing[package] + + def clear_deps(self): + """ + Clear dependencies without changing needed/optional. + + This is an ackward way to have a separate section to recommend + a package after system main dependencies. + + TODO: rework the logic to prevent needing it. + """ + + self.missing = {} + self.missing_pkg = {} + + def check_missing(self, progs): + """ + Update self.missing_pkg, using progs dict to convert from the + agnostic package name to distro-specific one. + + Returns an string with the packages to be installed, sorted and + with eventual duplicates removed. + """ + + self.missing_pkg = {} + + for prog, dtype in sorted(self.missing.items()): + # At least on some LTS distros like CentOS 7, texlive doesn't + # provide all packages we need. When such distros are + # detected, we have to disable PDF output. + # + # So, we need to ignore the packages that distros would + # need for LaTeX to work + if DepManager.is_pdf(dtype) and not self.pdf: + self.optional -= 1 + continue + + if not dtype in self.missing_pkg: + self.missing_pkg[dtype] = [] + + self.missing_pkg[dtype].append(progs.get(prog, prog)) + + install = [] + for dtype, pkgs in self.missing_pkg.items(): + install += pkgs + + return " ".join(sorted(set(install))) + + def warn_install(self): + """ + Emit warnings/errors related to missing packages. + """ + + output_msg = "" + + for dtype in sorted(self.missing_pkg.keys()): + progs = " ".join(sorted(set(self.missing_pkg[dtype]))) + + try: + name = DepManager.name(dtype) + output_msg += f'{name}:\t{progs}\n' + except KeyError: + raise KeyError(f"ERROR!!!: invalid dtype for {progs}: {dtype}") + + if output_msg: + print(f"\n{output_msg}") + +class AncillaryMethods: + """ + Ancillary methods that checks for missing dependencies for different + types of types, like binaries, python modules, rpm deps, etc. + """ + + @staticmethod + def which(prog): + """ + Our own implementation of which(). We could instead use + shutil.which(), but this function is simple enough. + Probably faster to use this implementation than to import shutil. + """ + for path in os.environ.get("PATH", "").split(":"): + full_path = os.path.join(path, prog) + if os.access(full_path, os.X_OK): + return full_path + + return None + + @staticmethod + def run(*args, **kwargs): + """ + Excecute a command, hiding its output by default. + Preserve compatibility with older Python versions. + """ + + capture_output = kwargs.pop('capture_output', False) + + if capture_output: + if 'stdout' not in kwargs: + kwargs['stdout'] = subprocess.PIPE + if 'stderr' not in kwargs: + kwargs['stderr'] = subprocess.PIPE + else: + if 'stdout' not in kwargs: + kwargs['stdout'] = subprocess.DEVNULL + if 'stderr' not in kwargs: + kwargs['stderr'] = subprocess.DEVNULL + + # Don't break with older Python versions + if 'text' in kwargs and sys.version_info < (3, 7): + kwargs['universal_newlines'] = kwargs.pop('text') + + return subprocess.run(*args, **kwargs) + +class MissingCheckers(AncillaryMethods): + """ + Contains some ancillary checkers for different types of binaries and + package managers. + """ + + def __init__(self, args, texlive): + """ + Initialize its internal variables + """ + self.pdf = args.pdf + self.virtualenv = args.virtualenv + self.version_check = args.version_check + self.texlive = texlive + + self.min_version = (0, 0, 0) + self.cur_version = (0, 0, 0) + + self.deps = DepManager(self.pdf) + + self.need_symlink = 0 + self.need_sphinx = 0 + + self.verbose_warn_install = 1 + + self.virtenv_dir = "" + self.install = "" + self.python_cmd = "" + + self.virtenv_prefix = ["sphinx_", "Sphinx_" ] + + def check_missing_file(self, files, package, dtype): + """ + Does the file exists? If not, add it to missing dependencies. + """ + for f in files: + if os.path.exists(f): + return + self.deps.add_package(package, dtype) + + def check_program(self, prog, dtype): + """ + Does the program exists and it is at the PATH? + If not, add it to missing dependencies. + """ + found = self.which(prog) + if found: + return found + + self.deps.add_package(prog, dtype) + + return None + + def check_perl_module(self, prog, dtype): + """ + Does perl have a dependency? Is it available? + If not, add it to missing dependencies. + + Right now, we still need Perl for doc build, as it is required + by some tools called at docs or kernel build time, like: + + tools/docs/documentation-file-ref-check + + Also, checkpatch is on Perl. + """ + + # While testing with lxc download template, one of the + # distros (Oracle) didn't have perl - nor even an option to install + # before installing oraclelinux-release-el9 package. + # + # Check it before running an error. If perl is not there, + # add it as a mandatory package, as some parts of the doc builder + # needs it. + if not self.which("perl"): + self.deps.add_package("perl", DepManager.SYSTEM_MANDATORY) + self.deps.add_package(prog, dtype) + return + + try: + self.run(["perl", f"-M{prog}", "-e", "1"], check=True) + except subprocess.CalledProcessError: + self.deps.add_package(prog, dtype) + + def check_python_module(self, module, is_optional=False): + """ + Does a python module exists outside venv? If not, add it to missing + dependencies. + """ + if is_optional: + dtype = DepManager.PYTHON_OPTIONAL + else: + dtype = DepManager.PYTHON_MANDATORY + + try: + self.run([self.python_cmd, "-c", f"import {module}"], check=True) + except subprocess.CalledProcessError: + self.deps.add_package(module, dtype) + + def check_rpm_missing(self, pkgs, dtype): + """ + Does a rpm package exists? If not, add it to missing dependencies. + """ + for prog in pkgs: + try: + self.run(["rpm", "-q", prog], check=True) + except subprocess.CalledProcessError: + self.deps.add_package(prog, dtype) + + def check_pacman_missing(self, pkgs, dtype): + """ + Does a pacman package exists? If not, add it to missing dependencies. + """ + for prog in pkgs: + try: + self.run(["pacman", "-Q", prog], check=True) + except subprocess.CalledProcessError: + self.deps.add_package(prog, dtype) + + def check_missing_tex(self, is_optional=False): + """ + Does a LaTeX package exists? If not, add it to missing dependencies. + """ + if is_optional: + dtype = DepManager.PDF_OPTIONAL + else: + dtype = DepManager.PDF_MANDATORY + + kpsewhich = self.which("kpsewhich") + for prog, package in self.texlive.items(): + + # If kpsewhich is not there, just add it to deps + if not kpsewhich: + self.deps.add_package(package, dtype) + continue + + # Check if the package is needed + try: + result = self.run( + [kpsewhich, prog], stdout=subprocess.PIPE, text=True, check=True + ) + + # Didn't find. Add it + if not result.stdout.strip(): + self.deps.add_package(package, dtype) + + except subprocess.CalledProcessError: + # kpsewhich returned an error. Add it, just in case + self.deps.add_package(package, dtype) + + def get_sphinx_fname(self): + """ + Gets the binary filename for sphinx-build. + """ + if "SPHINXBUILD" in os.environ: + return os.environ["SPHINXBUILD"] + + fname = "sphinx-build" + if self.which(fname): + return fname + + fname = "sphinx-build-3" + if self.which(fname): + self.need_symlink = 1 + return fname + + return "" + + def get_sphinx_version(self, cmd): + """ + Gets sphinx-build version. + """ + env = os.environ.copy() + + # The sphinx-build tool has a bug: internally, it tries to set + # locale with locale.setlocale(locale.LC_ALL, ''). This causes a + # crash if language is not set. Detect and fix it. + try: + locale.setlocale(locale.LC_ALL, '') + except Exception: + env["LC_ALL"] = "C" + env["LANG"] = "C" + + try: + result = self.run([cmd, "--version"], env=env, + stdout=subprocess.PIPE, + stderr=subprocess.STDOUT, + text=True, check=True) + except (subprocess.CalledProcessError, FileNotFoundError): + return None + + for line in result.stdout.split("\n"): + match = re.match(r"^sphinx-build\s+([\d\.]+)(?:\+(?:/[\da-f]+)|b\d+)?\s*$", line) + if match: + return PythonVersion.parse_version(match.group(1)) + + match = re.match(r"^Sphinx.*\s+([\d\.]+)\s*$", line) + if match: + return PythonVersion.parse_version(match.group(1)) + + def check_sphinx(self, conf): + """ + Checks Sphinx minimal requirements + """ + try: + with open(conf, "r", encoding="utf-8") as f: + for line in f: + match = re.match(r"^\s*needs_sphinx\s*=\s*[\'\"]([\d\.]+)[\'\"]", line) + if match: + self.min_version = PythonVersion.parse_version(match.group(1)) + break + except IOError: + sys.exit(f"Can't open {conf}") + + if not self.min_version: + sys.exit(f"Can't get needs_sphinx version from {conf}") + + self.virtenv_dir = self.virtenv_prefix[0] + "latest" + + sphinx = self.get_sphinx_fname() + if not sphinx: + self.need_sphinx = 1 + return + + self.cur_version = self.get_sphinx_version(sphinx) + if not self.cur_version: + sys.exit(f"{sphinx} didn't return its version") + + if self.cur_version < self.min_version: + curver = PythonVersion.ver_str(self.cur_version) + minver = PythonVersion.ver_str(self.min_version) + + print(f"ERROR: Sphinx version is {curver}. It should be >= {minver}") + self.need_sphinx = 1 + return + + # On version check mode, just assume Sphinx has all mandatory deps + if self.version_check and self.cur_version >= RECOMMENDED_VERSION: + sys.exit(0) + + def catcheck(self, filename): + """ + Reads a file if it exists, returning as string. + If not found, returns an empty string. + """ + if os.path.exists(filename): + with open(filename, "r", encoding="utf-8") as f: + return f.read().strip() + return "" + + def get_system_release(self): + """ + Determine the system type. There's no unique way that would work + with all distros with a minimal package install. So, several + methods are used here. + + By default, it will use lsb_release function. If not available, it will + fail back to reading the known different places where the distro name + is stored. + + Several modern distros now have /etc/os-release, which usually have + a decent coverage. + """ + + system_release = "" + + if self.which("lsb_release"): + result = self.run(["lsb_release", "-d"], capture_output=True, text=True) + system_release = result.stdout.replace("Description:", "").strip() + + release_files = [ + "/etc/system-release", + "/etc/redhat-release", + "/etc/lsb-release", + "/etc/gentoo-release", + ] + + if not system_release: + for f in release_files: + system_release = self.catcheck(f) + if system_release: + break + + # This seems more common than LSB these days + if not system_release: + os_var = {} + try: + with open("/etc/os-release", "r", encoding="utf-8") as f: + for line in f: + match = re.match(r"^([\w\d\_]+)=\"?([^\"]*)\"?\n", line) + if match: + os_var[match.group(1)] = match.group(2) + + system_release = os_var.get("NAME", "") + if "VERSION_ID" in os_var: + system_release += " " + os_var["VERSION_ID"] + elif "VERSION" in os_var: + system_release += " " + os_var["VERSION"] + except IOError: + pass + + if not system_release: + system_release = self.catcheck("/etc/issue") + + system_release = system_release.strip() + + return system_release + +class SphinxDependencyChecker(MissingCheckers): + """ + Main class for checking Sphinx documentation build dependencies. + + - Check for missing system packages; + - Check for missing Python modules; + - Check for missing LaTeX packages needed by PDF generation; + - Propose Sphinx install via Python Virtual environment; + - Propose Sphinx install via distro-specific package install. + """ + def __init__(self, args): + """Initialize checker variables""" + + # List of required texlive packages on Fedora and OpenSuse + texlive = { + "amsfonts.sty": "texlive-amsfonts", + "amsmath.sty": "texlive-amsmath", + "amssymb.sty": "texlive-amsfonts", + "amsthm.sty": "texlive-amscls", + "anyfontsize.sty": "texlive-anyfontsize", + "atbegshi.sty": "texlive-oberdiek", + "bm.sty": "texlive-tools", + "capt-of.sty": "texlive-capt-of", + "cmap.sty": "texlive-cmap", + "ctexhook.sty": "texlive-ctex", + "ecrm1000.tfm": "texlive-ec", + "eqparbox.sty": "texlive-eqparbox", + "eu1enc.def": "texlive-euenc", + "fancybox.sty": "texlive-fancybox", + "fancyvrb.sty": "texlive-fancyvrb", + "float.sty": "texlive-float", + "fncychap.sty": "texlive-fncychap", + "footnote.sty": "texlive-mdwtools", + "framed.sty": "texlive-framed", + "luatex85.sty": "texlive-luatex85", + "multirow.sty": "texlive-multirow", + "needspace.sty": "texlive-needspace", + "palatino.sty": "texlive-psnfss", + "parskip.sty": "texlive-parskip", + "polyglossia.sty": "texlive-polyglossia", + "tabulary.sty": "texlive-tabulary", + "threeparttable.sty": "texlive-threeparttable", + "titlesec.sty": "texlive-titlesec", + "ucs.sty": "texlive-ucs", + "upquote.sty": "texlive-upquote", + "wrapfig.sty": "texlive-wrapfig", + } + + super().__init__(args, texlive) + + self.need_pip = False + self.rec_sphinx_upgrade = 0 + + self.system_release = self.get_system_release() + self.activate_cmd = "" + + # Some distros may not have a Sphinx shipped package compatible with + # our minimal requirements + self.package_supported = True + + # Recommend a new python version + self.recommend_python = None + + # Certain hints are meant to be shown only once + self.distro_msg = None + + self.latest_avail_ver = (0, 0, 0) + self.venv_ver = (0, 0, 0) + + prefix = os.environ.get("srctree", ".") + "/" + + self.conf = prefix + "Documentation/conf.py" + self.requirement_file = prefix + "Documentation/sphinx/requirements.txt" + + def get_install_progs(self, progs, cmd, extra=None): + """ + Check for missing dependencies using the provided program mapping. + + The actual distro-specific programs are mapped via progs argument. + """ + install = self.deps.check_missing(progs) + + if self.verbose_warn_install: + self.deps.warn_install() + + if not install: + return + + if cmd: + if self.verbose_warn_install: + msg = "You should run:" + else: + msg = "" + + if extra: + msg += "\n\t" + extra.replace("\n", "\n\t") + + return(msg + "\n\tsudo " + cmd + " " + install) + + return None + + # + # Distro-specific hints methods + # + + def give_debian_hints(self): + """ + Provide package installation hints for Debian-based distros. + """ + progs = { + "Pod::Usage": "perl-modules", + "convert": "imagemagick", + "dot": "graphviz", + "ensurepip": "python3-venv", + "python-sphinx": "python3-sphinx", + "rsvg-convert": "librsvg2-bin", + "virtualenv": "virtualenv", + "xelatex": "texlive-xetex", + "yaml": "python3-yaml", + } + + if self.pdf: + pdf_pkgs = { + "fonts-dejavu": [ + "/usr/share/fonts/truetype/dejavu/DejaVuSans.ttf", + ], + "fonts-noto-cjk": [ + "/usr/share/fonts/noto-cjk/NotoSansCJK-Regular.ttc", + "/usr/share/fonts/opentype/noto/NotoSansCJK-Regular.ttc", + "/usr/share/fonts/opentype/noto/NotoSerifCJK-Regular.ttc", + ], + "tex-gyre": [ + "/usr/share/texmf/tex/latex/tex-gyre/tgtermes.sty" + ], + "texlive-fonts-recommended": [ + "/usr/share/texlive/texmf-dist/fonts/tfm/adobe/zapfding/pzdr.tfm", + ], + "texlive-lang-chinese": [ + "/usr/share/texlive/texmf-dist/tex/latex/ctex/ctexhook.sty", + ], + } + + for package, files in pdf_pkgs.items(): + self.check_missing_file(files, package, DepManager.PDF_MANDATORY) + + self.check_program("dvipng", DepManager.PDF_MANDATORY) + + if not self.distro_msg: + self.distro_msg = \ + "Note: ImageMagick is broken on some distros, affecting PDF output. For more details:\n" \ + "\thttps://askubuntu.com/questions/1158894/imagemagick-still-broken-using-with-usr-bin-convert" + + return self.get_install_progs(progs, "apt-get install") + + def give_redhat_hints(self): + """ + Provide package installation hints for RedHat-based distros + (Fedora, RHEL and RHEL-based variants). + """ + progs = { + "Pod::Usage": "perl-Pod-Usage", + "convert": "ImageMagick", + "dot": "graphviz", + "python-sphinx": "python3-sphinx", + "rsvg-convert": "librsvg2-tools", + "virtualenv": "python3-virtualenv", + "xelatex": "texlive-xetex-bin", + "yaml": "python3-pyyaml", + } + + fedora_tex_pkgs = [ + "dejavu-sans-fonts", + "dejavu-sans-mono-fonts", + "dejavu-serif-fonts", + "texlive-collection-fontsrecommended", + "texlive-collection-latex", + "texlive-xecjk", + ] + + fedora = False + rel = None + + match = re.search(r"(release|Linux)\s+(\d+)", self.system_release) + if match: + rel = int(match.group(2)) + + if not rel: + print("Couldn't identify release number") + noto_sans_redhat = None + self.pdf = False + elif re.search("Fedora", self.system_release): + # Fedora 38 and upper use this CJK font + + noto_sans_redhat = "google-noto-sans-cjk-fonts" + fedora = True + else: + # Almalinux, CentOS, RHEL, ... + + # at least up to version 9 (and Fedora < 38), that's the CJK font + noto_sans_redhat = "google-noto-sans-cjk-ttc-fonts" + + progs["virtualenv"] = "python-virtualenv" + + if not rel or rel < 8: + print("ERROR: Distro not supported. Too old?") + return + + # RHEL 8 uses Python 3.6, which is not compatible with + # the build system anymore. Suggest Python 3.11 + if rel == 8: + self.check_program("python3.9", DepManager.SYSTEM_MANDATORY) + progs["python3.9"] = "python39" + progs["yaml"] = "python39-pyyaml" + + self.recommend_python = True + + # There's no python39-sphinx package. Only pip is supported + self.package_supported = False + + if not self.distro_msg: + self.distro_msg = \ + "Note: RHEL-based distros typically require extra repositories.\n" \ + "For most, enabling epel and crb are enough:\n" \ + "\tsudo dnf install -y epel-release\n" \ + "\tsudo dnf config-manager --set-enabled crb\n" \ + "Yet, some may have other required repositories. Those commands could be useful:\n" \ + "\tsudo dnf repolist all\n" \ + "\tsudo dnf repoquery --available --info <pkgs>\n" \ + "\tsudo dnf config-manager --set-enabled '*' # enable all - probably not what you want" + + if self.pdf: + pdf_pkgs = [ + "/usr/share/fonts/google-noto-cjk/NotoSansCJK-Regular.ttc", + "/usr/share/fonts/google-noto-sans-cjk-fonts/NotoSansCJK-Regular.ttc", + ] + + self.check_missing_file(pdf_pkgs, noto_sans_redhat, DepManager.PDF_MANDATORY) + + self.check_rpm_missing(fedora_tex_pkgs, DepManager.PDF_MANDATORY) + + self.check_missing_tex(DepManager.PDF_MANDATORY) + + # There's no texlive-ctex on RHEL 8 repositories. This will + # likely affect CJK pdf build only. + if not fedora and rel == 8: + self.deps.del_package("texlive-ctex") + + return self.get_install_progs(progs, "dnf install") + + def give_opensuse_hints(self): + """ + Provide package installation hints for openSUSE-based distros + (Leap and Tumbleweed). + """ + progs = { + "Pod::Usage": "perl-Pod-Usage", + "convert": "ImageMagick", + "dot": "graphviz", + "python-sphinx": "python3-sphinx", + "virtualenv": "python3-virtualenv", + "xelatex": "texlive-xetex-bin texlive-dejavu", + "yaml": "python3-pyyaml", + } + + suse_tex_pkgs = [ + "texlive-babel-english", + "texlive-caption", + "texlive-colortbl", + "texlive-courier", + "texlive-dvips", + "texlive-helvetic", + "texlive-makeindex", + "texlive-metafont", + "texlive-metapost", + "texlive-palatino", + "texlive-preview", + "texlive-times", + "texlive-zapfchan", + "texlive-zapfding", + ] + + progs["latexmk"] = "texlive-latexmk-bin" + + match = re.search(r"(Leap)\s+(\d+).(\d)", self.system_release) + if match: + rel = int(match.group(2)) + + # Leap 15.x uses Python 3.6, which is not compatible with + # the build system anymore. Suggest Python 3.11 + if rel == 15: + if not self.which(self.python_cmd): + self.check_program("python3.11", DepManager.SYSTEM_MANDATORY) + progs["python3.11"] = "python311" + self.recommend_python = True + + progs.update({ + "python-sphinx": "python311-Sphinx python311-Sphinx-latex", + "virtualenv": "python311-virtualenv", + "yaml": "python311-PyYAML", + }) + else: + # Tumbleweed defaults to Python 3.11 + + progs.update({ + "python-sphinx": "python313-Sphinx python313-Sphinx-latex", + "virtualenv": "python313-virtualenv", + "yaml": "python313-PyYAML", + }) + + # FIXME: add support for installing CJK fonts + # + # I tried hard, but was unable to find a way to install + # "Noto Sans CJK SC" on openSUSE + + if self.pdf: + self.check_rpm_missing(suse_tex_pkgs, DepManager.PDF_MANDATORY) + if self.pdf: + self.check_missing_tex() + + return self.get_install_progs(progs, "zypper install --no-recommends") + + def give_mageia_hints(self): + """ + Provide package installation hints for Mageia and OpenMandriva. + """ + progs = { + "Pod::Usage": "perl-Pod-Usage", + "convert": "ImageMagick", + "dot": "graphviz", + "python-sphinx": "python3-sphinx", + "rsvg-convert": "librsvg2", + "virtualenv": "python3-virtualenv", + "xelatex": "texlive", + "yaml": "python3-yaml", + } + + tex_pkgs = [ + "texlive-fontsextra", + "texlive-fonts-asian", + "fonts-ttf-dejavu", + ] + + if re.search(r"OpenMandriva", self.system_release): + packager_cmd = "dnf install" + noto_sans = "noto-sans-cjk-fonts" + tex_pkgs = [ + "texlive-collection-basic", + "texlive-collection-langcjk", + "texlive-collection-fontsextra", + "texlive-collection-fontsrecommended" + ] + + # Tested on OpenMandriva Lx 4.3 + progs["convert"] = "imagemagick" + progs["yaml"] = "python-pyyaml" + progs["python-virtualenv"] = "python-virtualenv" + progs["python-sphinx"] = "python-sphinx" + progs["xelatex"] = "texlive" + + self.check_program("python-virtualenv", DepManager.PYTHON_MANDATORY) + + # On my tests with openMandriva LX 4.0 docker image, upgraded + # to 4.3, python-virtualenv package is broken: it is missing + # ensurepip. Without it, the alternative would be to run: + # python3 -m venv --without-pip ~/sphinx_latest, but running + # pip there won't install sphinx at venv. + # + # Add a note about that. + + if not self.distro_msg: + self.distro_msg = \ + "Notes:\n"\ + "1. for venv, ensurepip could be broken, preventing its install method.\n" \ + "2. at least on OpenMandriva LX 4.3, texlive packages seem broken" + + else: + packager_cmd = "urpmi" + noto_sans = "google-noto-sans-cjk-ttc-fonts" + + progs["latexmk"] = "texlive-collection-basic" + + if self.pdf: + pdf_pkgs = [ + "/usr/share/fonts/google-noto-cjk/NotoSansCJK-Regular.ttc", + "/usr/share/fonts/TTF/NotoSans-Regular.ttf", + ] + + self.check_missing_file(pdf_pkgs, noto_sans, DepManager.PDF_MANDATORY) + self.check_rpm_missing(tex_pkgs, DepManager.PDF_MANDATORY) + + return self.get_install_progs(progs, packager_cmd) + + def give_arch_linux_hints(self): + """ + Provide package installation hints for ArchLinux. + """ + progs = { + "convert": "imagemagick", + "dot": "graphviz", + "latexmk": "texlive-core", + "rsvg-convert": "extra/librsvg", + "virtualenv": "python-virtualenv", + "xelatex": "texlive-xetex", + "yaml": "python-yaml", + } + + archlinux_tex_pkgs = [ + "texlive-basic", + "texlive-binextra", + "texlive-core", + "texlive-fontsrecommended", + "texlive-langchinese", + "texlive-langcjk", + "texlive-latexextra", + "ttf-dejavu", + ] + + if self.pdf: + self.check_pacman_missing(archlinux_tex_pkgs, + DepManager.PDF_MANDATORY) + + self.check_missing_file(["/usr/share/fonts/noto-cjk/NotoSansCJK-Regular.ttc"], + "noto-fonts-cjk", + DepManager.PDF_MANDATORY) + + + return self.get_install_progs(progs, "pacman -S") + + def give_gentoo_hints(self): + """ + Provide package installation hints for Gentoo. + """ + texlive_deps = [ + "dev-texlive/texlive-fontsrecommended", + "dev-texlive/texlive-latexextra", + "dev-texlive/texlive-xetex", + "media-fonts/dejavu", + ] + + progs = { + "convert": "media-gfx/imagemagick", + "dot": "media-gfx/graphviz", + "rsvg-convert": "gnome-base/librsvg", + "virtualenv": "dev-python/virtualenv", + "xelatex": " ".join(texlive_deps), + "yaml": "dev-python/pyyaml", + "python-sphinx": "dev-python/sphinx", + } + + if self.pdf: + pdf_pkgs = { + "media-fonts/dejavu": [ + "/usr/share/fonts/dejavu/DejaVuSans.ttf", + ], + "media-fonts/noto-cjk": [ + "/usr/share/fonts/noto-cjk/NotoSansCJKsc-Regular.otf", + "/usr/share/fonts/noto-cjk/NotoSerifCJK-Regular.ttc", + ], + } + for package, files in pdf_pkgs.items(): + self.check_missing_file(files, package, DepManager.PDF_MANDATORY) + + # Handling dependencies is a nightmare, as Gentoo refuses to emerge + # some packages if there's no package.use file describing them. + # To make it worse, compilation flags shall also be present there + # for some packages. If USE is not perfect, error/warning messages + # like those are shown: + # + # !!! The following binary packages have been ignored due to non matching USE: + # + # =media-gfx/graphviz-12.2.1-r1 X pdf -python_single_target_python3_13 qt6 svg + # =media-gfx/graphviz-12.2.1-r1 X pdf python_single_target_python3_12 -python_single_target_python3_13 qt6 svg + # =media-gfx/graphviz-12.2.1-r1 X pdf qt6 svg + # =media-gfx/graphviz-12.2.1-r1 X pdf -python_single_target_python3_10 qt6 svg + # =media-gfx/graphviz-12.2.1-r1 X pdf -python_single_target_python3_10 python_single_target_python3_12 -python_single_target_python3_13 qt6 svg + # =media-fonts/noto-cjk-20190416 X + # =app-text/texlive-core-2024-r1 X cjk -xetex + # =app-text/texlive-core-2024-r1 X -xetex + # =app-text/texlive-core-2024-r1 -xetex + # =dev-libs/zziplib-0.13.79-r1 sdl + # + # And will ignore such packages, installing the remaining ones. That + # affects mostly the image extension and PDF generation. + + # Package dependencies and the minimal needed args: + portages = { + "graphviz": "media-gfx/graphviz", + "imagemagick": "media-gfx/imagemagick", + "media-libs": "media-libs/harfbuzz icu", + "media-fonts": "media-fonts/noto-cjk", + "texlive": "app-text/texlive-core xetex", + "zziblib": "dev-libs/zziplib sdl", + } + + extra_cmds = "" + if not self.distro_msg: + self.distro_msg = "Note: Gentoo requires package.use to be adjusted before emerging packages" + + use_base = "/etc/portage/package.use" + files = glob(f"{use_base}/*") + + for fname, portage in portages.items(): + install = False + + while install is False: + if not files: + # No files under package.usage. Install all + install = True + break + + args = portage.split(" ") + + name = args.pop(0) + + cmd = ["grep", "-l", "-E", rf"^{name}\b" ] + files + result = self.run(cmd, stdout=subprocess.PIPE, text=True) + if result.returncode or not result.stdout.strip(): + # File containing portage name not found + install = True + break + + # Ensure that needed USE flags are present + if args: + match_fname = result.stdout.strip() + with open(match_fname, 'r', encoding='utf8', + errors='backslashreplace') as fp: + for line in fp: + for arg in args: + if arg.startswith("-"): + continue + + if not re.search(rf"\s*{arg}\b", line): + # Needed file argument not found + install = True + break + + # Everything looks ok, don't install + break + + # emit a code to setup missing USE + if install: + extra_cmds += (f"sudo su -c 'echo \"{portage}\" > {use_base}/{fname}'\n") + + # Now, we can use emerge and let it respect USE + return self.get_install_progs(progs, + "emerge --ask --changed-use --binpkg-respect-use=y", + extra_cmds) + + def get_install(self): + """ + OS-specific hints logic. Seeks for a hinter. If found, use it to + provide package-manager specific install commands. + + Otherwise, outputs install instructions for the meta-packages. + + Returns a string with the command to be executed to install the + the needed packages, if distro found. Otherwise, return just a + list of packages that require installation. + """ + os_hints = { + re.compile("Red Hat Enterprise Linux"): self.give_redhat_hints, + re.compile("Fedora"): self.give_redhat_hints, + re.compile("AlmaLinux"): self.give_redhat_hints, + re.compile("Amazon Linux"): self.give_redhat_hints, + re.compile("CentOS"): self.give_redhat_hints, + re.compile("openEuler"): self.give_redhat_hints, + re.compile("Oracle Linux Server"): self.give_redhat_hints, + re.compile("Rocky Linux"): self.give_redhat_hints, + re.compile("Springdale Open Enterprise"): self.give_redhat_hints, + + re.compile("Ubuntu"): self.give_debian_hints, + re.compile("Debian"): self.give_debian_hints, + re.compile("Devuan"): self.give_debian_hints, + re.compile("Kali"): self.give_debian_hints, + re.compile("Mint"): self.give_debian_hints, + + re.compile("openSUSE"): self.give_opensuse_hints, + + re.compile("Mageia"): self.give_mageia_hints, + re.compile("OpenMandriva"): self.give_mageia_hints, + + re.compile("Arch Linux"): self.give_arch_linux_hints, + re.compile("Gentoo"): self.give_gentoo_hints, + } + + # If the OS is detected, use per-OS hint logic + for regex, os_hint in os_hints.items(): + if regex.search(self.system_release): + return os_hint() + + # + # Fall-back to generic hint code for other distros + # That's far from ideal, specially for LaTeX dependencies. + # + progs = {"sphinx-build": "sphinx"} + if self.pdf: + self.check_missing_tex() + + self.distro_msg = \ + f"I don't know distro {self.system_release}.\n" \ + "So, I can't provide you a hint with the install procedure.\n" \ + "There are likely missing dependencies." + + return self.get_install_progs(progs, None) + + # + # Common dependencies + # + def deactivate_help(self): + """ + Print a helper message to disable a virtual environment. + """ + + print("\n If you want to exit the virtualenv, you can use:") + print("\tdeactivate") + + def get_virtenv(self): + """ + Give a hint about how to activate an already-existing virtual + environment containing sphinx-build. + + Returns a tuble with (activate_cmd_path, sphinx_version) with + the newest available virtual env. + """ + + cwd = os.getcwd() + + activates = [] + + # Add all sphinx prefixes with possible version numbers + for p in self.virtenv_prefix: + activates += glob(f"{cwd}/{p}[0-9]*/bin/activate") + + activates.sort(reverse=True, key=str.lower) + + # Place sphinx_latest first, if it exists + for p in self.virtenv_prefix: + activates = glob(f"{cwd}/{p}*latest/bin/activate") + activates + + ver = (0, 0, 0) + for f in activates: + # Discard too old Sphinx virtual environments + match = re.search(r"(\d+)\.(\d+)\.(\d+)", f) + if match: + ver = (int(match.group(1)), int(match.group(2)), int(match.group(3))) + + if ver < self.min_version: + continue + + sphinx_cmd = f.replace("activate", "sphinx-build") + if not os.path.isfile(sphinx_cmd): + continue + + ver = self.get_sphinx_version(sphinx_cmd) + + if not ver: + venv_dir = f.replace("/bin/activate", "") + print(f"Warning: virtual environment {venv_dir} is not working.\n" \ + "Python version upgrade? Remove it with:\n\n" \ + "\trm -rf {venv_dir}\n\n") + else: + if self.need_sphinx and ver >= self.min_version: + return (f, ver) + elif PythonVersion.parse_version(ver) > self.cur_version: + return (f, ver) + + return ("", ver) + + def recommend_sphinx_upgrade(self): + """ + Check if Sphinx needs to be upgraded. + + Returns a tuple with the higest available Sphinx version if found. + Otherwise, returns None to indicate either that no upgrade is needed + or no venv was found. + """ + + # Avoid running sphinx-builds from venv if cur_version is good + if self.cur_version and self.cur_version >= RECOMMENDED_VERSION: + self.latest_avail_ver = self.cur_version + return None + + # Get the highest version from sphinx_*/bin/sphinx-build and the + # corresponding command to activate the venv/virtenv + self.activate_cmd, self.venv_ver = self.get_virtenv() + + # Store the highest version from Sphinx existing virtualenvs + if self.activate_cmd and self.venv_ver > self.cur_version: + self.latest_avail_ver = self.venv_ver + else: + if self.cur_version: + self.latest_avail_ver = self.cur_version + else: + self.latest_avail_ver = (0, 0, 0) + + # As we don't know package version of Sphinx, and there's no + # virtual environments, don't check if upgrades are needed + if not self.virtualenv: + if not self.latest_avail_ver: + return None + + return self.latest_avail_ver + + # Either there are already a virtual env or a new one should be created + self.need_pip = True + + if not self.latest_avail_ver: + return None + + # Return if the reason is due to an upgrade or not + if self.latest_avail_ver != (0, 0, 0): + if self.latest_avail_ver < RECOMMENDED_VERSION: + self.rec_sphinx_upgrade = 1 + + return self.latest_avail_ver + + def recommend_package(self): + """ + Recommend installing Sphinx as a distro-specific package. + """ + + print("\n2) As a package with:") + + old_need = self.deps.need + old_optional = self.deps.optional + + self.pdf = False + self.deps.optional = 0 + old_verbose = self.verbose_warn_install + self.verbose_warn_install = 0 + + self.deps.clear_deps() + + self.deps.add_package("python-sphinx", DepManager.PYTHON_MANDATORY) + + cmd = self.get_install() + if cmd: + print(cmd) + + self.deps.need = old_need + self.deps.optional = old_optional + self.verbose_warn_install = old_verbose + + def recommend_sphinx_version(self, virtualenv_cmd): + """ + Provide recommendations for installing or upgrading Sphinx based + on current version. + + The logic here is complex, as it have to deal with different versions: + + - minimal supported version; + - minimal PDF version; + - recommended version. + + It also needs to work fine with both distro's package and + venv/virtualenv + """ + + if self.recommend_python: + cur_ver = sys.version_info[:3] + if cur_ver < MIN_PYTHON_VERSION: + print(f"\nPython version {cur_ver} is incompatible with doc build.\n" \ + "Please upgrade it and re-run.\n") + return + + # Version is OK. Nothing to do. + if self.cur_version != (0, 0, 0) and self.cur_version >= RECOMMENDED_VERSION: + return + + if self.latest_avail_ver: + latest_avail_ver = PythonVersion.ver_str(self.latest_avail_ver) + + if not self.need_sphinx: + # sphinx-build is present and its version is >= $min_version + + # only recommend enabling a newer virtenv version if makes sense. + if self.latest_avail_ver and self.latest_avail_ver > self.cur_version: + print(f"\nYou may also use the newer Sphinx version {latest_avail_ver} with:") + if f"{self.virtenv_prefix}" in os.getcwd(): + print("\tdeactivate") + print(f"\t. {self.activate_cmd}") + self.deactivate_help() + return + + if self.latest_avail_ver and self.latest_avail_ver >= RECOMMENDED_VERSION: + return + + if not self.virtualenv: + # No sphinx either via package or via virtenv. As we can't + # Compare the versions here, just return, recommending the + # user to install it from the package distro. + if not self.latest_avail_ver or self.latest_avail_ver == (0, 0, 0): + return + + # User doesn't want a virtenv recommendation, but he already + # installed one via virtenv with a newer version. + # So, print commands to enable it + if self.latest_avail_ver > self.cur_version: + print(f"\nYou may also use the Sphinx virtualenv version {latest_avail_ver} with:") + if f"{self.virtenv_prefix}" in os.getcwd(): + print("\tdeactivate") + print(f"\t. {self.activate_cmd}") + self.deactivate_help() + return + print("\n") + else: + if self.need_sphinx: + self.deps.need += 1 + + # Suggest newer versions if current ones are too old + if self.latest_avail_ver and self.latest_avail_ver >= self.min_version: + if self.latest_avail_ver >= RECOMMENDED_VERSION: + print(f"\nNeed to activate Sphinx (version {latest_avail_ver}) on virtualenv with:") + print(f"\t. {self.activate_cmd}") + self.deactivate_help() + return + + # Version is above the minimal required one, but may be + # below the recommended one. So, print warnings/notes + if self.latest_avail_ver < RECOMMENDED_VERSION: + print(f"Warning: It is recommended at least Sphinx version {RECOMMENDED_VERSION}.") + + # At this point, either it needs Sphinx or upgrade is recommended, + # both via pip + + if self.rec_sphinx_upgrade: + if not self.virtualenv: + print("Instead of install/upgrade Python Sphinx pkg, you could use pip/pypi with:\n\n") + else: + print("To upgrade Sphinx, use:\n\n") + else: + print("\nSphinx needs to be installed either:\n1) via pip/pypi with:\n") + + if not virtualenv_cmd: + print(" Currently not possible.\n") + print(" Please upgrade Python to a newer version and run this script again") + else: + print(f"\t{virtualenv_cmd} {self.virtenv_dir}") + print(f"\t. {self.virtenv_dir}/bin/activate") + print(f"\tpip install -r {self.requirement_file}") + self.deactivate_help() + + if self.package_supported: + self.recommend_package() + + print("\n" \ + " Please note that Sphinx currentlys produce false-positive\n" \ + " warnings when the same name is used for more than one type (functions,\n" \ + " structs, enums,...). This is known Sphinx bug. For more details, see:\n" \ + "\thttps://github.com/sphinx-doc/sphinx/pull/8313") + + def check_needs(self): + """ + Main method that checks needed dependencies and provides + recommendations. + """ + self.python_cmd = sys.executable + + # Check if Sphinx is already accessible from current environment + self.check_sphinx(self.conf) + + if self.system_release: + print(f"Detected OS: {self.system_release}.") + else: + print("Unknown OS") + if self.cur_version != (0, 0, 0): + ver = PythonVersion.ver_str(self.cur_version) + print(f"Sphinx version: {ver}\n") + + # Check the type of virtual env, depending on Python version + virtualenv_cmd = None + + if sys.version_info < MIN_PYTHON_VERSION: + min_ver = ver_str(MIN_PYTHON_VERSION) + print(f"ERROR: at least python {min_ver} is required to build the kernel docs") + self.need_sphinx = 1 + + self.venv_ver = self.recommend_sphinx_upgrade() + + if self.need_pip: + if sys.version_info < MIN_PYTHON_VERSION: + self.need_pip = False + print("Warning: python version is not supported.") + else: + virtualenv_cmd = f"{self.python_cmd} -m venv" + self.check_python_module("ensurepip") + + # Check for needed programs/tools + self.check_perl_module("Pod::Usage", DepManager.SYSTEM_MANDATORY) + + self.check_program("make", DepManager.SYSTEM_MANDATORY) + self.check_program("which", DepManager.SYSTEM_MANDATORY) + + self.check_program("dot", DepManager.SYSTEM_OPTIONAL) + self.check_program("convert", DepManager.SYSTEM_OPTIONAL) + + self.check_python_module("yaml") + + if self.pdf: + self.check_program("xelatex", DepManager.PDF_MANDATORY) + self.check_program("rsvg-convert", DepManager.PDF_MANDATORY) + self.check_program("latexmk", DepManager.PDF_MANDATORY) + + # Do distro-specific checks and output distro-install commands + cmd = self.get_install() + if cmd: + print(cmd) + + # If distro requires some special instructions, print here. + # Please notice that get_install() needs to be called first. + if self.distro_msg: + print("\n" + self.distro_msg) + + if not self.python_cmd: + if self.need == 1: + sys.exit("Can't build as 1 mandatory dependency is missing") + elif self.need: + sys.exit(f"Can't build as {self.need} mandatory dependencies are missing") + + # Check if sphinx-build is called sphinx-build-3 + if self.need_symlink: + sphinx_path = self.which("sphinx-build-3") + if sphinx_path: + print(f"\tsudo ln -sf {sphinx_path} /usr/bin/sphinx-build\n") + + self.recommend_sphinx_version(virtualenv_cmd) + print("") + + if not self.deps.optional: + print("All optional dependencies are met.") + + if self.deps.need == 1: + sys.exit("Can't build as 1 mandatory dependency is missing") + elif self.deps.need: + sys.exit(f"Can't build as {self.deps.need} mandatory dependencies are missing") + + print("Needed package dependencies are met.") + +DESCRIPTION = """ +Process some flags related to Sphinx installation and documentation build. +""" + + +def main(): + """Main function""" + parser = argparse.ArgumentParser(description=DESCRIPTION) + + parser.add_argument( + "--no-virtualenv", + action="store_false", + dest="virtualenv", + help="Recommend installing Sphinx instead of using a virtualenv", + ) + + parser.add_argument( + "--no-pdf", + action="store_false", + dest="pdf", + help="Don't check for dependencies required to build PDF docs", + ) + + parser.add_argument( + "--version-check", + action="store_true", + dest="version_check", + help="If version is compatible, don't check for missing dependencies", + ) + + args = parser.parse_args() + + checker = SphinxDependencyChecker(args) + + PythonVersion.check_python(MIN_PYTHON_VERSION, + bail_out=True, success_on_error=True) + checker.check_needs() + +# Call main if not used as module +if __name__ == "__main__": + main() diff --git a/tools/docs/test_doc_build.py b/tools/docs/test_doc_build.py new file mode 100755 index 000000000000..47b4606569f9 --- /dev/null +++ b/tools/docs/test_doc_build.py @@ -0,0 +1,513 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0 +# Copyright(c) 2025: Mauro Carvalho Chehab <mchehab+huawei@kernel.org> +# +# pylint: disable=R0903,R0912,R0913,R0914,R0917,C0301 + +""" +Install minimal supported requirements for different Sphinx versions +and optionally test the build. +""" + +import argparse +import asyncio +import os.path +import shutil +import sys +import time +import subprocess + +# Minimal python version supported by the building system. + +PYTHON = os.path.basename(sys.executable) + +min_python_bin = None + +for i in range(9, 13): + p = f"python3.{i}" + if shutil.which(p): + min_python_bin = p + break + +if not min_python_bin: + min_python_bin = PYTHON + +# Starting from 8.0, Python 3.9 is not supported anymore. +PYTHON_VER_CHANGES = {(8, 0, 0): PYTHON} + +DEFAULT_VERSIONS_TO_TEST = [ + (3, 4, 3), # Minimal supported version + (5, 3, 0), # CentOS Stream 9 / AlmaLinux 9 + (6, 1, 1), # Debian 12 + (7, 2, 1), # openSUSE Leap 15.6 + (7, 2, 6), # Ubuntu 24.04 LTS + (7, 4, 7), # Ubuntu 24.10 + (7, 3, 0), # openSUSE Tumbleweed + (8, 1, 3), # Fedora 42 + (8, 2, 3) # Latest version - covers rolling distros +] + +# Sphinx versions to be installed and their incremental requirements +SPHINX_REQUIREMENTS = { + # Oldest versions we support for each package required by Sphinx 3.4.3 + (3, 4, 3): { + "docutils": "0.16", + "alabaster": "0.7.12", + "babel": "2.8.0", + "certifi": "2020.6.20", + "docutils": "0.16", + "idna": "2.10", + "imagesize": "1.2.0", + "Jinja2": "2.11.2", + "MarkupSafe": "1.1.1", + "packaging": "20.4", + "Pygments": "2.6.1", + "PyYAML": "5.1", + "requests": "2.24.0", + "snowballstemmer": "2.0.0", + "sphinxcontrib-applehelp": "1.0.2", + "sphinxcontrib-devhelp": "1.0.2", + "sphinxcontrib-htmlhelp": "1.0.3", + "sphinxcontrib-jsmath": "1.0.1", + "sphinxcontrib-qthelp": "1.0.3", + "sphinxcontrib-serializinghtml": "1.1.4", + "urllib3": "1.25.9", + }, + + # Update package dependencies to a more modern base. The goal here + # is to avoid to many incremental changes for the next entries + (3, 5, 0): { + "alabaster": "0.7.13", + "babel": "2.17.0", + "certifi": "2025.6.15", + "idna": "3.10", + "imagesize": "1.4.1", + "packaging": "25.0", + "Pygments": "2.8.1", + "requests": "2.32.4", + "snowballstemmer": "3.0.1", + "sphinxcontrib-applehelp": "1.0.4", + "sphinxcontrib-htmlhelp": "2.0.1", + "sphinxcontrib-serializinghtml": "1.1.5", + "urllib3": "2.0.0", + }, + + # Starting from here, ensure all docutils versions are covered with + # supported Sphinx versions. Other packages are upgraded only when + # required by pip + (4, 0, 0): { + "PyYAML": "5.1", + }, + (4, 1, 0): { + "docutils": "0.17", + "Pygments": "2.19.1", + "Jinja2": "3.0.3", + "MarkupSafe": "2.0", + }, + (4, 3, 0): {}, + (4, 4, 0): {}, + (4, 5, 0): { + "docutils": "0.17.1", + }, + (5, 0, 0): {}, + (5, 1, 0): {}, + (5, 2, 0): { + "docutils": "0.18", + "Jinja2": "3.1.2", + "MarkupSafe": "2.0", + "PyYAML": "5.3.1", + }, + (5, 3, 0): { + "docutils": "0.18.1", + }, + (6, 0, 0): {}, + (6, 1, 0): {}, + (6, 2, 0): { + "PyYAML": "5.4.1", + }, + (7, 0, 0): {}, + (7, 1, 0): {}, + (7, 2, 0): { + "docutils": "0.19", + "PyYAML": "6.0.1", + "sphinxcontrib-serializinghtml": "1.1.9", + }, + (7, 2, 6): { + "docutils": "0.20", + }, + (7, 3, 0): { + "alabaster": "0.7.14", + "PyYAML": "6.0.1", + "tomli": "2.0.1", + }, + (7, 4, 0): { + "docutils": "0.20.1", + "PyYAML": "6.0.1", + }, + (8, 0, 0): { + "docutils": "0.21", + }, + (8, 1, 0): { + "docutils": "0.21.1", + "PyYAML": "6.0.1", + "sphinxcontrib-applehelp": "1.0.7", + "sphinxcontrib-devhelp": "1.0.6", + "sphinxcontrib-htmlhelp": "2.0.6", + "sphinxcontrib-qthelp": "1.0.6", + }, + (8, 2, 0): { + "docutils": "0.21.2", + "PyYAML": "6.0.1", + "sphinxcontrib-serializinghtml": "1.1.9", + }, +} + + +class AsyncCommands: + """Excecute command synchronously""" + + def __init__(self, fp=None): + + self.stdout = None + self.stderr = None + self.output = None + self.fp = fp + + def log(self, out, verbose, is_info=True): + out = out.removesuffix('\n') + + if verbose: + if is_info: + print(out) + else: + print(out, file=sys.stderr) + + if self.fp: + self.fp.write(out + "\n") + + async def _read(self, stream, verbose, is_info): + """Ancillary routine to capture while displaying""" + + while stream is not None: + line = await stream.readline() + if line: + out = line.decode("utf-8", errors="backslashreplace") + self.log(out, verbose, is_info) + if is_info: + self.stdout += out + else: + self.stderr += out + else: + break + + async def run(self, cmd, capture_output=False, check=False, + env=None, verbose=True): + + """ + Execute an arbitrary command, handling errors. + + Please notice that this class is not thread safe + """ + + self.stdout = "" + self.stderr = "" + + self.log("$ " + " ".join(cmd), verbose) + + proc = await asyncio.create_subprocess_exec(cmd[0], + *cmd[1:], + env=env, + stdout=asyncio.subprocess.PIPE, + stderr=asyncio.subprocess.PIPE) + + # Handle input and output in realtime + await asyncio.gather( + self._read(proc.stdout, verbose, True), + self._read(proc.stderr, verbose, False), + ) + + await proc.wait() + + if check and proc.returncode > 0: + raise subprocess.CalledProcessError(returncode=proc.returncode, + cmd=" ".join(cmd), + output=self.stdout, + stderr=self.stderr) + + if capture_output: + if proc.returncode > 0: + self.log(f"Error {proc.returncode}", verbose=True, is_info=False) + return "" + + return self.output + + ret = subprocess.CompletedProcess(args=cmd, + returncode=proc.returncode, + stdout=self.stdout, + stderr=self.stderr) + + return ret + + +class SphinxVenv: + """ + Installs Sphinx on one virtual env per Sphinx version with a minimal + set of dependencies, adjusting them to each specific version. + """ + + def __init__(self): + """Initialize instance variables""" + + self.built_time = {} + self.first_run = True + + async def _handle_version(self, args, fp, + cur_ver, cur_requirements, python_bin): + """Handle a single Sphinx version""" + + cmd = AsyncCommands(fp) + + ver = ".".join(map(str, cur_ver)) + + if not self.first_run and args.wait_input and args.build: + ret = input("Press Enter to continue or 'a' to abort: ").strip().lower() + if ret == "a": + print("Aborted.") + sys.exit() + else: + self.first_run = False + + venv_dir = f"Sphinx_{ver}" + req_file = f"requirements_{ver}.txt" + + cmd.log(f"\nSphinx {ver} with {python_bin}", verbose=True) + + # Create venv + await cmd.run([python_bin, "-m", "venv", venv_dir], + verbose=args.verbose, check=True) + pip = os.path.join(venv_dir, "bin/pip") + + # Create install list + reqs = [] + for pkg, verstr in cur_requirements.items(): + reqs.append(f"{pkg}=={verstr}") + + reqs.append(f"Sphinx=={ver}") + + await cmd.run([pip, "install"] + reqs, check=True, verbose=args.verbose) + + # Freeze environment + result = await cmd.run([pip, "freeze"], verbose=False, check=True) + + # Pip install succeeded. Write requirements file + if args.req_file: + with open(req_file, "w", encoding="utf-8") as fp: + fp.write(result.stdout) + + if args.build: + start_time = time.time() + + # Prepare a venv environment + env = os.environ.copy() + bin_dir = os.path.join(venv_dir, "bin") + env["PATH"] = bin_dir + ":" + env["PATH"] + env["VIRTUAL_ENV"] = venv_dir + if "PYTHONHOME" in env: + del env["PYTHONHOME"] + + # Test doc build + await cmd.run(["make", "cleandocs"], env=env, check=True) + make = ["make"] + + if args.output: + sphinx_build = os.path.realpath(f"{bin_dir}/sphinx-build") + make += [f"O={args.output}", f"SPHINXBUILD={sphinx_build}"] + + if args.make_args: + make += args.make_args + + make += args.targets + + if args.verbose: + cmd.log(f". {bin_dir}/activate", verbose=True) + await cmd.run(make, env=env, check=True, verbose=True) + if args.verbose: + cmd.log("deactivate", verbose=True) + + end_time = time.time() + elapsed_time = end_time - start_time + hours, minutes = divmod(elapsed_time, 3600) + minutes, seconds = divmod(minutes, 60) + + hours = int(hours) + minutes = int(minutes) + seconds = int(seconds) + + self.built_time[ver] = f"{hours:02d}:{minutes:02d}:{seconds:02d}" + + cmd.log(f"Finished doc build for Sphinx {ver}. Elapsed time: {self.built_time[ver]}", verbose=True) + + async def run(self, args): + """ + Navigate though multiple Sphinx versions, handling each of them + on a loop. + """ + + if args.log: + fp = open(args.log, "w", encoding="utf-8") + if not args.verbose: + args.verbose = False + else: + fp = None + if not args.verbose: + args.verbose = True + + cur_requirements = {} + python_bin = min_python_bin + + vers = set(SPHINX_REQUIREMENTS.keys()) | set(args.versions) + + for cur_ver in sorted(vers): + if cur_ver in SPHINX_REQUIREMENTS: + new_reqs = SPHINX_REQUIREMENTS[cur_ver] + cur_requirements.update(new_reqs) + + if cur_ver in PYTHON_VER_CHANGES: # pylint: disable=R1715 + python_bin = PYTHON_VER_CHANGES[cur_ver] + + if cur_ver not in args.versions: + continue + + if args.min_version: + if cur_ver < args.min_version: + continue + + if args.max_version: + if cur_ver > args.max_version: + break + + await self._handle_version(args, fp, cur_ver, cur_requirements, + python_bin) + + if args.build: + cmd = AsyncCommands(fp) + cmd.log("\nSummary:", verbose=True) + for ver, elapsed_time in sorted(self.built_time.items()): + cmd.log(f"\tSphinx {ver} elapsed time: {elapsed_time}", + verbose=True) + + if fp: + fp.close() + +def parse_version(ver_str): + """Convert a version string into a tuple.""" + + return tuple(map(int, ver_str.split("."))) + + +DEFAULT_VERS = " - " +DEFAULT_VERS += "\n - ".join(map(lambda v: f"{v[0]}.{v[1]}.{v[2]}", + DEFAULT_VERSIONS_TO_TEST)) + +SCRIPT = os.path.relpath(__file__) + +DESCRIPTION = f""" +This tool allows creating Python virtual environments for different +Sphinx versions that are supported by the Linux Kernel build system. + +Besides creating the virtual environment, it can also test building +the documentation using "make htmldocs" (and/or other doc targets). + +If called without "--versions" argument, it covers the versions shipped +on major distros, plus the lowest supported version: + +{DEFAULT_VERS} + +A typical usage is to run: + + {SCRIPT} -m -l sphinx_builds.log + +This will create one virtual env for the default version set and run +"make htmldocs" for each version, creating a log file with the +excecuted commands on it. + +NOTE: The build time can be very long, specially on old versions. Also, there +is a known bug with Sphinx version 6.0.x: each subprocess uses a lot of +memory. That, together with "-jauto" may cause OOM killer to cause +failures at the doc generation. To minimize the risk, you may use the +"-a" command line parameter to constrain the built directories and/or +reduce the number of threads from "-jauto" to, for instance, "-j4": + + {SCRIPT} -m -V 6.0.1 -a "SPHINXDIRS=process" "SPHINXOPTS='-j4'" + +""" + +MAKE_TARGETS = [ + "htmldocs", + "texinfodocs", + "infodocs", + "latexdocs", + "pdfdocs", + "epubdocs", + "xmldocs", +] + +async def main(): + """Main program""" + + parser = argparse.ArgumentParser(description=DESCRIPTION, + formatter_class=argparse.RawDescriptionHelpFormatter) + + ver_group = parser.add_argument_group("Version range options") + + ver_group.add_argument('-V', '--versions', nargs="*", + default=DEFAULT_VERSIONS_TO_TEST,type=parse_version, + help='Sphinx versions to test') + ver_group.add_argument('--min-version', "--min", type=parse_version, + help='Sphinx minimal version') + ver_group.add_argument('--max-version', "--max", type=parse_version, + help='Sphinx maximum version') + ver_group.add_argument('-f', '--full', action='store_true', + help='Add all Sphinx (major,minor) supported versions to the version range') + + build_group = parser.add_argument_group("Build options") + + build_group.add_argument('-b', '--build', action='store_true', + help='Build documentation') + build_group.add_argument('-a', '--make-args', nargs="*", + help='extra arguments for make, like SPHINXDIRS=netlink/specs', + ) + build_group.add_argument('-t', '--targets', nargs="+", choices=MAKE_TARGETS, + default=[MAKE_TARGETS[0]], + help="make build targets. Default: htmldocs.") + build_group.add_argument("-o", '--output', + help="output directory for the make O=OUTPUT") + + other_group = parser.add_argument_group("Other options") + + other_group.add_argument('-r', '--req-file', action='store_true', + help='write a requirements.txt file') + other_group.add_argument('-l', '--log', + help='Log command output on a file') + other_group.add_argument('-v', '--verbose', action='store_true', + help='Verbose all commands') + other_group.add_argument('-i', '--wait-input', action='store_true', + help='Wait for an enter before going to the next version') + + args = parser.parse_args() + + if not args.make_args: + args.make_args = [] + + sphinx_versions = sorted(list(SPHINX_REQUIREMENTS.keys())) + + if args.full: + args.versions += list(SPHINX_REQUIREMENTS.keys()) + + venv = SphinxVenv() + await venv.run(args) + + +# Call main method +if __name__ == "__main__": + asyncio.run(main()) |