Various contributor script changes.

Update `msgfmt.py` to recent python 3 version.
Instruct CMake to load `Gettext.Tools` version `0.21.0.1` when using NuGet.
Cleanup `regen_POT_file.sh`.
Correct list of languages in `LanguageNames.txt`
Cleanup various locale Perl scripts.
Make `find-PO-no-Id.sh` use glob file access.
Cleanup output of `locale_diagnostics.sh`
Cleanup bash scripts and add double-quoting to prevent string issues.
Move various locale scripts to `locale/scripts` directory.

Signed-off-by: Emily Mabrey <emabrey@tenacityaudio.org>

images/scripts/degimpify.sh

@@ -12,9 +12,9 @@ shopt -s nullglob
 input_dirs="$*"
 
 if [ $# -eq 0 ]; then
-    input_dirs[0]="./EditButtons"
-    input_dirs[1]="./TranscriptionImages"
-    input_dirs[2]="./ControlButtons"
+    input_dirs[0]="${BASH_SOURCE[0]}/../../images/EditButtons"
+    input_dirs[1]="${BASH_SOURCE[0]}/../../images/TranscriptionImages"
+    input_dirs[2]="${BASH_SOURCE[0]}/../../images/ControlButtons"
 fi
 
 for dir in "${input_dirs[@]}"

locale/CMakeLists.txt

@@ -16,9 +16,9 @@ if( GETTEXT_FOUND )
    mark_as_advanced( FORCE GETTEXT_MSGMERGE_EXECUTABLE )
    set( msgfmt "${GETTEXT_MSGFMT_EXECUTABLE}" )
 elseif( PYTHON )
-   set( msgfmt "${PYTHON}" "${TARGET_ROOT}/msgfmt.py" )
+   set( msgfmt "${PYTHON}" "${TARGET_ROOT}/scripts/msgfmt.py" )
 elseif( CMAKE_SYSTEM_NAME MATCHES "Windows" )
-   nuget_package( pkgdir "Gettext.Tools" "0.20.1.1" )
+   nuget_package( pkgdir "Gettext.Tools" "0.21.0.1" )
    file( TO_NATIVE_PATH "${pkgdir}/tools/bin/msgfmt.exe" msgfmt )
 endif()
 
@@ -80,9 +80,8 @@ endforeach()
 add_custom_target( "${TARGET}" ALL DEPENDS ${OUTPUTS} SOURCES "${SOURCES}" )
 
 if( NOT CMAKE_SYSTEM_NAME MATCHES "Darwin" )
-   if( NOT WIN32)
-      install( DIRECTORY ${locale}/
-               TYPE LOCALE )
+   if( NOT WIN32 )
+      install( DIRECTORY ${locale} TYPE LOCALE )
    endif()
 endif()
 

locale/regen_POT_file.sh

@@ -1,56 +0,0 @@
-#!/bin/sh
-# Run this script with locale as the current directory
-set -o errexit
-echo ";; Recreating tenacity.pot using .h, .cpp and .mm files"
-for path in ../modules/mod-* ../libraries/lib-* ../include ../src ; do
-   find $path -name \*.h -o -name \*.cpp -o -name \*.mm
-done | LANG=c sort | \
-sed -E 's/\.\.\///g' |\
-xargs xgettext \
---no-wrap \
---default-domain=tenacity \
---directory=.. \
---keyword=_ --keyword=XO --keyword=XC:1,2c --keyword=XXO --keyword=XXC:1,2c --keyword=XP:1,2 --keyword=XPC:1,2,4c \
---add-comments=" i18n" \
---add-location=file  \
---copyright-holder='Tenacity Contributors' \
---package-name="tenacity" \
---package-version='3.0.4' \
---msgid-bugs-address="emabrey@tenacityaudio.org" \
---add-location=file -L C -o tenacity.pot
-echo ";; Adding nyquist files to tenacity.pot"
-for path in ../plug-ins ; do find $path -name \*.ny -not -name rms.ny; done | LANG=c sort | \
-sed -E 's/\.\.\///g' |\
-xargs xgettext \
---no-wrap \
---default-domain=tenacity \
---directory=.. \
---keyword=_ --keyword=_C:1,2c --keyword=ngettext:1,2 --keyword=ngettextc:1,2,4c \
---add-comments=" i18n" \
---add-location=file  \
---copyright-holder='Tenacity Contributors' \
---package-name="tenacity" \
---package-version='3.0.4' \
---msgid-bugs-address="emabrey@tenacityaudio.org" \
---add-location=file -L Lisp -j -o tenacity.pot
-#Disable PO file modification because it will hurt Weblate merge -- emabrey.
-#if test "${TENACITY_ONLY_POT:-}" = 'y'; then
-#    return 0
-#fi
-#echo ";; Updating the .po files - Updating Project-Id-Version"
-#for i in *.po; do
-#    sed -e '/^"Project-Id-Version:/c\
-#    "Project-Id-Version: tenacity 3.0.4\\n"' $i > TEMP; mv TEMP $i
-#done
-#echo ";; Updating the .po files"
-#sed 's/.*/echo "msgmerge --lang=& &.po tenacity.pot -o &.po";\
-#msgmerge --no-wrap --lang=& &.po tenacity.pot -o &.po;/g' LINGUAS | bash
-#echo ";; Removing '#~|' (which confuse Windows version of msgcat)"
-#for i in *.po; do
-#    sed '/^#~|/d' $i > TEMP; mv TEMP $i
-#done
-#echo ""
-echo ";; POT file generated"
-echo ""
-head -n 11 tenacity.pot | tail -n 3
-wc -l tenacity.pot

locale/scripts/LanguageNames.txt

@@ -5,7 +5,7 @@ bg Български
 bn বাংলা
 bs Bosanski
 ca Català
-ca_ES@valencia Valencià
+ca@valencia Valencià
 co Corsu
 cs Čeština
 cy Cymraeg
@@ -15,7 +15,6 @@ el Ελληνικά
 en English
 es Español
 eu Euskara
-eu_ES Euskara (Espainiako)
 fa فارسی
 fi Suomi
 fr Français
@@ -40,19 +39,19 @@ nb Norsk
 nl Nederlands
 oc Occitan
 pl Polski
-pt Português
+pt_PT Português
 pt_BR Português (Brasil)
 ro Română
 ru Русский
 sk Slovenčina
 sl Slovenščina
-sr_RS Српски
-sr_RS@latin Srpski
+sr Српски
+sr_Latn Srpski
 sv Svenska
 ta தமிழ்
 tg Тоҷикӣ
 tr Türkçe
 uk Українська
 vi Tiếng Việt
-zh_CN 中文（简体）
-zh_TW 中文（繁體）
+zh 中文（简体）
+zh_Hant 中文（繁體）

locale/scripts/find-PO-no-Id.sh

@@ -0,0 +1,18 @@
+#!/bin/bash
+
+# Run this script on a directory to handle PO files with no id
+
+shopt -s nullglob
+
+input_dirs="$*"
+
+if [ $# -eq 0 ]; then
+    input_dirs[0]="${BASH_SOURCE[0]}/../"
+fi
+
+for dir in "${input_dirs[@]}"
+do
+    for po_file in "${dir}"/*.po; do
+       sed -i '/^Project/d' "$po_file"
+    done
+done

locale/scripts/find-POT-duplicates.pl

@@ -2,8 +2,7 @@
 
 use strict 'vars';
 
-open(my $fh, "<", "audacity.pot")
-    or die "Can't open < audacity.pot: $!";
+open(my $fh, "<", "../tenacity.pot") or die "Can't open < ../tenacity.pot: $!";
 
 my @sourcelines = ();
 

locale/scripts/fix-PO-files.py

@@ -1,7 +1,7 @@
 import polib
 import os
 
-localedir = './'
+localedir = '../'
 count_seen = 0
 count_modified = 0
 

locale/scripts/locale_diagnostics.sh

@@ -0,0 +1,23 @@
+#!/bin/bash
+
+# Report how complete each translation catalog is
+
+# How many messages in total?
+total=$(grep -c '^msgid' "${BASH_SOURCE[0]}/../tenacity.pot")
+
+
+declare -i badlines
+for po_file in "${BASH_SOURCE[0]}/../"*.po; do
+    # If there are errors from msgcmp, then the last line on standard error
+    # contains a count of problematic messages; else it won't match the
+    # pattern in awk, so assume no errors
+    errors=$(msgcmp "$po_file" "${BASH_SOURCE[0]}/../tenacity.pot" 2>&1 | awk '/msgcmp: found [0-9]* fatal error/ { nn = $3 } END {print 0+nn}')
+    complete=$((total-errors))
+
+    # detect whether this sequence occurs in any .po file.  It will break msgfmt on Windows.
+    badlines=$(grep -F -c '#~|' "${po_file}")
+
+    echo "$po_file: completed $complete of $total ($((complete*100/total))%); Windows Incompatible Lines: $badlines"
+done | sort -n -t , -k3
+
+exit 0

locale/scripts/msgfmt.py

@@ -5,8 +5,7 @@
 
 This program converts a textual Uniforum-style message catalog (.po file) into
 a binary GNU catalog (.mo file).  This is essentially the same function as the
-GNU msgfmt program, however, it is a simpler implementation.  Currently it
-does not handle plural forms but it does handle message contexts.
+GNU msgfmt program, however, it is a simpler implementation.
 
 Usage: msgfmt.py [OPTIONS] filename.po
 
@@ -33,11 +32,12 @@ import struct
 import array
 from email.parser import HeaderParser
 
-__version__ = "1.2"
+__version__ = "1.1"
 
 MESSAGES = {}
 
 
+
 def usage(code, msg=''):
     print(__doc__, file=sys.stderr)
     if msg:
@@ -45,16 +45,15 @@ def usage(code, msg=''):
     sys.exit(code)
 
 
-def add(ctxt, id, str, fuzzy):
+
+def add(id, str, fuzzy):
     "Add a non-fuzzy translation to the dictionary."
     global MESSAGES
     if not fuzzy and str:
-        if ctxt is None:
-            MESSAGES[id] = str
-        else:
-            MESSAGES[b"%b\x04%b" % (ctxt, id)] = str
+        MESSAGES[id] = str
 
 
+
 def generate():
     "Return the generated output."
     global MESSAGES
@@ -96,10 +95,10 @@ def generate():
     return output
 
 
+
 def make(filename, outfile):
     ID = 1
     STR = 2
-    CTXT = 3
 
     # Compute .mo name from .po name and arguments
     if filename.endswith('.po'):
@@ -116,7 +115,7 @@ def make(filename, outfile):
         print(msg, file=sys.stderr)
         sys.exit(1)
 
-    section = msgctxt = None
+    section = None
     fuzzy = 0
 
     # Start off assuming Latin-1, so everything decodes without failure,
@@ -130,8 +129,8 @@ def make(filename, outfile):
         lno += 1
         # If we get a comment line after a msgstr, this is a new entry
         if l[0] == '#' and section == STR:
-            add(msgctxt, msgid, msgstr, fuzzy)
-            section = msgctxt = None
+            add(msgid, msgstr, fuzzy)
+            section = None
             fuzzy = 0
         # Record a fuzzy mark
         if l[:2] == '#,' and 'fuzzy' in l:
@@ -139,16 +138,10 @@ def make(filename, outfile):
         # Skip comments
         if l[0] == '#':
             continue
-        # Now we are in a msgid or msgctxt section, output previous section
-        if l.startswith('msgctxt'):
+        # Now we are in a msgid section, output previous section
+        if l.startswith('msgid') and not l.startswith('msgid_plural'):
             if section == STR:
-                add(msgctxt, msgid, msgstr, fuzzy)
-            section = CTXT
-            l = l[7:]
-            msgctxt = b''
-        elif l.startswith('msgid') and not l.startswith('msgid_plural'):
-            if section == STR:
-                add(msgctxt, msgid, msgstr, fuzzy)
+                add(msgid, msgstr, fuzzy)
                 if not msgid:
                     # See whether there is an encoding declaration
                     p = HeaderParser()
@@ -190,9 +183,7 @@ def make(filename, outfile):
         if not l:
             continue
         l = ast.literal_eval(l)
-        if section == CTXT:
-            msgctxt += l.encode(encoding)
-        elif section == ID:
+        if section == ID:
             msgid += l.encode(encoding)
         elif section == STR:
             msgstr += l.encode(encoding)
@@ -203,7 +194,7 @@ def make(filename, outfile):
             sys.exit(1)
     # Add last entry
     if section == STR:
-        add(msgctxt, msgid, msgstr, fuzzy)
+        add(msgid, msgstr, fuzzy)
 
     # Compute output
     output = generate()
@@ -215,6 +206,7 @@ def make(filename, outfile):
         print(msg, file=sys.stderr)
 
 
+
 def main():
     try:
         opts, args = getopt.getopt(sys.argv[1:], 'hVo:',

locale/scripts/regen_POT_file.sh

@@ -0,0 +1,44 @@
+#!/bin/bash
+# Run this script with /locale/script/ as the current directory
+set -o errexit
+
+echo ";; Recreating tenacity.pot using .h, .cpp and .mm files"
+for path in "${BASH_SOURCE[0]}/../modules/mod-*" "${BASH_SOURCE[0]}/../libraries/lib-*" "${BASH_SOURCE[0]}/../include" "${BASH_SOURCE[0]}/../src" ; do
+   find "$path" -name \*.h -o -name \*.cpp -o -name \*.mm
+done | LANG=c sort | \
+sed -E 's/\.\.\///g' |\
+xargs xgettext \
+--no-wrap \
+--default-domain=tenacity \
+--directory="${BASH_SOURCE[0]}/../.." \
+--keyword=_ --keyword=XO --keyword=XC:1,2c --keyword=XXO --keyword=XXC:1,2c --keyword=XP:1,2 --keyword=XPC:1,2,4c \
+--add-comments=" i18n" \
+--add-location=file  \
+--copyright-holder='Tenacity Contributors' \
+--package-name="tenacity" \
+--package-version='3.0.4' \
+--msgid-bugs-address="emabrey@tenacityaudio.org" \
+--add-location=file -L C -o "${BASH_SOURCE[0]}/../tenacity.pot"
+
+echo ";; Adding nyquist files to tenacity.pot"
+for path in "${BASH_SOURCE[0]}/../../plug-ins/"* ; do find "$path" -name \*.ny -not -name rms.ny; done | LANG=c sort | \
+sed -E 's/\.\.\///g' |\
+xargs xgettext \
+--no-wrap \
+--default-domain=tenacity \
+--directory="${BASH_SOURCE[0]}/../.." \
+--keyword=_ --keyword=_C:1,2c --keyword=ngettext:1,2 --keyword=ngettextc:1,2,4c \
+--add-comments=" i18n" \
+--add-location=file  \
+--copyright-holder='Tenacity Contributors' \
+--package-name="tenacity" \
+--package-version='3.0.4' \
+--msgid-bugs-address="emabrey@tenacityaudio.org" \
+--add-location=file -L Lisp -j -o "${BASH_SOURCE[0]}/../tenacity.pot"
+
+echo ""
+echo ";; POT file generated"
+echo ""
+
+head -n 11 "${BASH_SOURCE[0]}/../tenacity.pot" | tail -n 3
+wc -l "${BASH_SOURCE[0]}/../tenacity.pot"

locale/update_po_files.sh

@@ -1,55 +0,0 @@
-#!/bin/sh
-# Run this script with locale as the current directory
-set -o errexit
-echo ";; Recreating tenacity.pot using .h, .cpp and .mm files"
-for path in ../modules/mod-* ../libraries/lib-* ../include ../src ; do
-   find $path -name \*.h -o -name \*.cpp -o -name \*.mm
-done | LANG=c sort | \
-sed -E 's/\.\.\///g' |\
-xargs xgettext \
---no-wrap \
---default-domain=tenacity \
---directory=.. \
---keyword=_ --keyword=XO --keyword=XC:1,2c --keyword=XXO --keyword=XXC:1,2c --keyword=XP:1,2 --keyword=XPC:1,2,4c \
---add-comments=" i18n" \
---add-location=file  \
---copyright-holder='Tenacity Contributors' \
---package-name="tenacity" \
---package-version='3.0.4' \
---msgid-bugs-address="emabrey@tenacityaudio.org" \
---add-location=file -L C -o tenacity.pot
-echo ";; Adding nyquist files to tenacity.pot"
-for path in ../plug-ins ; do find $path -name \*.ny -not -name rms.ny; done | LANG=c sort | \
-sed -E 's/\.\.\///g' |\
-xargs xgettext \
---no-wrap \
---default-domain=tenacity \
---directory=.. \
---keyword=_ --keyword=_C:1,2c --keyword=ngettext:1,2 --keyword=ngettextc:1,2,4c \
---add-comments=" i18n" \
---add-location=file  \
---copyright-holder='Tenacity Contributors' \
---package-name="tenacity" \
---package-version='3.0.4' \
---msgid-bugs-address="emabrey@tenacityaudio.org" \
---add-location=file -L Lisp -j -o tenacity.pot
-if test "${TENACITY_ONLY_POT:-}" = 'y'; then
-    return 0
-fi
-echo ";; Updating the .po files - Updating Project-Id-Version"
-for i in *.po; do
-    sed -e '/^"Project-Id-Version:/c\
-    "Project-Id-Version: tenacity 3.0.4\\n"' $i > TEMP; mv TEMP $i
-done
-echo ";; Updating the .po files"
-sed 's/.*/echo "msgmerge --lang=& &.po tenacity.pot -o &.po";\
-msgmerge --no-wrap --lang=& &.po tenacity.pot -o &.po;/g' LINGUAS | bash
-echo ";; Removing '#~|' (which confuse Windows version of msgcat)"
-for i in *.po; do
-    sed '/^#~|/d' $i > TEMP; mv TEMP $i
-done
-echo ""
-echo ";;Translation updated"
-echo ""
-head -n 11 tenacity.pot | tail -n 3
-wc -l tenacity.pot