From d91ad2976c1735654fa881b0a44dc625d23c1d25 Mon Sep 17 00:00:00 2001
From: Fabian Boehm <FHomborg@gmail.com>
Date: Tue, 5 Mar 2024 17:24:17 +0100
Subject: [PATCH] Make fish_xgettext sorta work with rust

This is absolutely disgusting code, but it works out okay-ish.

The problem is xgettext has no rust support (it's stuck in review
limbo). So we use cargo-expand to extract all invocations of
gettext, and massage all that to generate a
messages.pot ourselves.

We also assume any string constant could be translated.
---
 build_tools/fish_xgettext.fish | 69 ++++++++++++++++++----------------
 1 file changed, 37 insertions(+), 32 deletions(-)

diff --git a/build_tools/fish_xgettext.fish b/build_tools/fish_xgettext.fish
index 6977d7c3f..77b8e9d17 100755
--- a/build_tools/fish_xgettext.fish
+++ b/build_tools/fish_xgettext.fish
@@ -1,18 +1,40 @@
 #!/usr/bin/env fish
 #
 # Tool to generate messages.pot
-# Extended to replace the old Makefile rule which did not port easily to CMake
 
-# This script was originally motivated to work around a quirk (or bug depending on your viewpoint)
-# of the xgettext command. See https://lists.gnu.org/archive/html/bug-gettext/2014-11/msg00006.html.
-# However, it turns out that even if that quirk did not exist we would still need something like
-# this script to properly extract descriptions. That's because we need to normalize the strings to
-# a format that xgettext will handle correctly. Also, `xgettext -LShell` doesn't correctly extract
-# all the strings we want translated. So we extract and normalize all such strings into a format
-# that `xgettext` can handle.
+# Create temporary directory for these operations. OS X `mktemp` is somewhat restricted, so this block
+# works around that - based on share/functions/funced.fish.
+set -q TMPDIR
+or set -l TMPDIR /tmp
+set -l tmpdir (mktemp -d $TMPDIR/fish.XXXXXX)
+or exit 1
 
-# Start with the C++ source
-xgettext -k -k_ -kN_ -LC++ --no-wrap -o messages.pot src/*.cpp src/*.h
+# This is a gigantic crime.
+# xgettext still does not support rust *at all*, so we use cargo-expand to get all our wgettext invocations.
+set -l expanded (cargo expand --lib; for f in fish{,_indent,_key_reader}; cargo expand --bin $f; end)
+
+# Extract any gettext call
+set -l strs (printf '%s\n' $expanded | grep -A1 wgettext_static_str |
+             grep 'widestring::internals::core::primitive::str =' |
+             string match -rg '"(.*)"' | string match -rv '^%ls$|^$' |
+             # escaping difference between gettext and cargo-expand: single-quotes
+             string replace -a "\'" "'" | sort -u)
+
+# Extract any constants
+set -a strs (string match -rv 'BUILD_VERSION:|PACKAGE_NAME' -- $expanded |
+             string match -rg 'const [A-Z_]*: &str = "(.*)"' | string replace -a "\'" "'")
+
+# We construct messages.pot ourselves instead of forcing this into msgmerge or whatever.
+# The escaping so far works out okay.
+for str in $strs
+    # grep -P needed for string escape to be compatible (PCRE-style),
+    # -H gives the filename, -n the line number.
+    # If you want to run this on non-GNU grep: Don't.
+    echo "#:" (grep -PHn -r -- (string escape --style=regex -- $str) src/ |
+    head -n1 | string replace -r ':\s.*' '')
+    echo "msgid \"$str\""
+    echo 'msgstr ""'
+end >messages.pot
 
 # This regex handles descriptions for `complete` and `function` statements. These messages are not
 # particularly important to translate. Hence the "implicit" label.
@@ -22,39 +44,22 @@ set -l implicit_regex '(?:^| +)(?:complete|function).*? (?:-d|--description) (([
 # than messages which should be implicitly translated.
 set -l explicit_regex '.*\( *_ (([\'"]).+?(?<!\\\\)\\2) *\).*'
 
-# Create temporary directory for these operations. OS X `mktemp` is somewhat restricted, so this block
-# works around that - based on share/functions/funced.fish.
-set -q TMPDIR
-or set -l TMPDIR /tmp
-set -l tmpdir (mktemp -d $TMPDIR/fish.XXXXXX)
-or exit 1
-
 mkdir -p $tmpdir/implicit/share/completions $tmpdir/implicit/share/functions
 mkdir -p $tmpdir/explicit/share/completions $tmpdir/explicit/share/functions
 
 for f in share/config.fish share/completions/*.fish share/functions/*.fish
     # Extract explicit attempts to translate a message. That is, those that are of the form
     # `(_ "message")`.
-    string replace --filter --regex $explicit_regex 'echo $1' <$f | fish >$tmpdir/explicit/$f.tmp 2>/dev/null
-    while read description
-        echo 'N_ "'(string replace --all '"' '\\"' -- $description)'"'
-    end <$tmpdir/explicit/$f.tmp >$tmpdir/explicit/$f
-    rm $tmpdir/explicit/$f.tmp
+    string replace --filter --regex $explicit_regex '$1' <$f | string unescape \
+        | string replace --all '"' '\\"' | string replace -r '(.*)' 'N_ "$1"' >$tmpdir/explicit/$f
 
     # Handle `complete` / `function` description messages. The `| fish` is subtle. It basically
     # avoids the need to use `source` with a command substitution that could affect the current
     # shell.
-    string replace --filter --regex $implicit_regex 'echo $1' <$f | fish >$tmpdir/implicit/$f.tmp 2>/dev/null
-    while read description
-        # We don't use `string escape` as shown in the next comment because it produces output that
-        # is not parsed correctly by xgettext. Instead just escape double-quotes and quote the
-        # resulting string.
-        echo 'N_ "'(string replace --all '"' '\\"' -- $description)'"'
-    end <$tmpdir/implicit/$f.tmp >$tmpdir/implicit/$f
-    rm $tmpdir/implicit/$f.tmp
+    string replace --filter --regex $implicit_regex '$1' <$f | string unescape \
+        | string replace --all '"' '\\"' | string replace -r '(.*)' 'N_ "$1"' >$tmpdir/implicit/$f
 end
 
-xgettext -j -k -kN_ -LShell --from-code=UTF-8 -cDescription --no-wrap -o messages.pot $tmpdir/explicit/share/*/*.fish
-xgettext -j -k -kN_ -LShell --from-code=UTF-8 -cDescription --no-wrap -o messages.pot $tmpdir/implicit/share/*/*.fish
+xgettext -j -k -kN_ -LShell --from-code=UTF-8 -cDescription --no-wrap -o messages.pot $tmpdir/{ex,im}plicit/share/*/*.fish
 
 rm -r $tmpdir