From 3e29793d046ad3e5a79dfe1ec75e9df3f2d87090 Mon Sep 17 00:00:00 2001
From: Kurtis Rader <krader@skepticism.us>
Date: Sat, 27 May 2017 22:13:18 -0700
Subject: [PATCH] improve detection of msgs to be translated

This change does several things. First, it works around a quirk of the
`xgetttext` command that only recognizes description strings in even
numbered position on the command. Second, it allows descriptions
introduced by the `-d` short flag to be recognized.

More importantly, it normalizes the strings so that `xgettext` correctly
extracts them into the *.po file. Prior to this change many fish script
strings were ignored due to how they were written (e.g., single versus
double quotes).

Fixes #4073
---
 Makefile.in                    | 15 +++++++-----
 build_tools/fish_xgettext.fish | 42 ++++++++++++++++++++++++++++++++++
 2 files changed, 51 insertions(+), 6 deletions(-)
 create mode 100755 build_tools/fish_xgettext.fish

diff --git a/Makefile.in b/Makefile.in
index 8577fed4d..3e5a7dd04 100644
--- a/Makefile.in
+++ b/Makefile.in
@@ -516,13 +516,12 @@ doc.h: $(HDR_FILES)
 		-e "s,@prefix\@,$(prefix),g" \
 		-e "s,@fish_build_version\@,$(FISH_BUILD_VERSION),g" \
 
-
 #
-# Compile translation files to binary format
+# Compile translation files to binary format.
 #
 %.gmo:
 	@echo "  msgfmt   $(em)$@$(sgr0)"
-	$v msgfmt -o $@ $*.po
+	$v msgfmt --use-fuzzy -o $@ $*.po
 
 #
 # Update existing po file or copy messages.pot
@@ -538,10 +537,14 @@ doc.h: $(HDR_FILES)
 #
 # Create a template translation object
 #
-messages.pot: $(wildcard src/*.cpp src/*.h share/completions/*.fish share/functions/*.fish)
+messages.pot: fish $(wildcard src/*.cpp src/*.h share/completions/*.fish share/functions/*.fish)
 	@echo "  xgettext $(em)$@$(sgr0)"
-	xgettext -k_ -kN_ $(wildcard src/*.cpp src/*.h) -o messages.pot
-	$v xgettext -j -k_ -kN_ -k--description -LShell --from-code=UTF-8 $(wildcard share/completions/*.fish share/functions/*.fish) share/config.fish -o messages.pot
+	$v xgettext -k -k_ -kN_ -LC++ --no-wrap -o messages.pot src/*.cpp src/*.h
+	$v rm -rf /tmp/fish
+	$v ./fish build_tools/fish_xgettext.fish
+	$v xgettext -j -k -kN_ -LShell --from-code=UTF-8 -cDescription --no-wrap -o messages.pot /tmp/fish/explicit/share/*.fish /tmp/fish/explicit/share/*/*.fish
+	$v xgettext -j -k -kN_ -LShell --from-code=UTF-8 -cDescription --no-wrap -o messages.pot /tmp/fish/implicit/share/*.fish /tmp/fish/implicit/share/*/*.fish
+	$v rm -rf /tmp/fish
 
 ifdef EXTRA_PCRE2
 src/builtin_string.cpp: $(PCRE2_H)
diff --git a/build_tools/fish_xgettext.fish b/build_tools/fish_xgettext.fish
new file mode 100755
index 000000000..eeeddaf32
--- /dev/null
+++ b/build_tools/fish_xgettext.fish
@@ -0,0 +1,42 @@
+#!/usr/bin/env fish
+#
+# 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.
+
+# This regex handles descriptions for `complete` and `function` statements. These messages are not
+# particularly important to translate. Hence the "implicit" label.
+set implicit_regex '(?:^| +)(?:complete|function) .*? (?:-d|--description) (([\'"]).+?(?<!\\\\)\\2).*'
+
+# This regex handles explicit requests to translate a message. These are more important to translate
+# than messages which should be implicitly translated.
+set explicit_regex '.*\( *_ (([\'"]).+?(?<!\\\\)\\2) *\).*'
+
+mkdir -p /tmp/fish/implicit/share/completions /tmp/fish/implicit/share/functions
+mkdir -p /tmp/fish/explicit/share/completions /tmp/fish/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 >/tmp/fish/explicit/$f.tmp ^/dev/null
+    while read description
+        echo 'N_ "'(string replace --all '"' '\\"' -- $description)'"'
+    end </tmp/fish/explicit/$f.tmp >/tmp/fish/explicit/$f
+    rm /tmp/fish/explicit/$f.tmp
+
+    # Handle `complete` / `function` description messages. The `| fish` is subtle. It basically
+    # avoids the need to use `source` with a command substituion that could affect the current
+    # shell.
+    string replace --filter --regex $implicit_regex 'echo $1' <$f | fish >/tmp/fish/implicit/$f.tmp ^/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 '"' '\\"' -- $description)'"'
+    end </tmp/fish/implicit/$f.tmp >/tmp/fish/implicit/$f
+    rm /tmp/fish/implicit/$f.tmp
+end