make libiast usable with C++

.appveyor.yml

@@ -1,16 +0,0 @@
-version: '{branch}-{build}'
-
-platform: x64
-
-environment:
-  CYG_ROOT: C:\cygwin64
-  CYG_SHELL: C:\cygwin64\bin\sh
-
-install:
-  - cmd: '%CYG_ROOT%\setup-x86_64.exe --quiet-mode --no-shortcuts --upgrade-also --packages autoconf,make,git,gcc,check'
-
-build_script:
-  - cmd: '%CYG_SHELL% -lc "cd $APPVEYOR_BUILD_FOLDER && sh autogen.sh && sh configure && make all"'
-
-test_script:
-  - cmd: '%CYG_SHELL% -lc "cd $APPVEYOR_BUILD_FOLDER && make test"'

.github/workflows/test.yml

@@ -0,0 +1,22 @@
+name: C/C++ CI
+
+on:
+  push:
+    branches: [ "master" ]
+  pull_request:
+    branches: [ "master" ]
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+
+    steps:
+    - uses: actions/checkout@v3
+    - name: dependencies
+      run: sudo apt update && sudo apt install -y autoconf make gcc lcov check pkg-config
+    - name: configure
+      run: sh autogen.sh && ./configure
+    - name: make
+      run: make
+    - name: maketest
+      run: make test

.gitignore

@@ -3,6 +3,7 @@
 *.d
 *.so
 *.so.*
+*.pc
 /iast
 /tests/test
 /tests/coverage

Makefile.in

@@ -19,13 +19,14 @@ INSTALL_DATA    = @INSTALL_DATA@
 INSTALL_PROGRAM = @INSTALL_PROGRAM@
 MKDIR_P         = @MKDIR_P@
 
-prefix      = @prefix@
-exec_prefix = @exec_prefix@
-bindir      = @bindir@
-libdir      = @libdir@
-datarootdir = @datarootdir@
-includedir  = @includedir@
-mandir      = @mandir@
+prefix       = @prefix@
+exec_prefix  = @exec_prefix@
+bindir       = @bindir@
+libdir       = @libdir@
+datarootdir  = @datarootdir@
+includedir   = @includedir@
+pkgconfigdir = @pkgconfigdir@
+mandir       = @mandir@
 
 USE_GCOV = @USE_GCOV@
 
@@ -57,7 +58,7 @@ $(ANAME): $(OBJECTS)
 	$(QUIET_AR) $(AR) rcs $@ $^
 
 $(REALNAME): $(OBJECTS)
-	$(QUIET_LD) $(CC) $^ -shared -o $@ $(LFLAGS)
+	$(QUIET_LD) $(CC) $^ -shared -Wl,-soname=$(SONAME) -o $@ $(LFLAGS)
 
 $(SONAME): $(REALNAME)
 	$(QUIET_LN) $(LN_S) -f $< $@
@@ -83,7 +84,7 @@ tests/test: $(OBJECTS) $(TEST_OBJECTS)
 install: install-prog install-lib install-dev
 
 install-prog:
-	$(MKDIR_P) $(bindir) $(libdir) $(mandir)/man1
+	$(MKDIR_P) $(bindir) $(mandir)/man1
 	$(INSTALL_PROGRAM) $(PROGNAME) $(bindir)
 	$(INSTALL_DATA) $(PROGNAME).1.gz $(mandir)/man1
 
@@ -92,9 +93,10 @@ install-lib:
 	cp -P $(REALNAME) $(SONAME) $(libdir)
 
 install-dev:
-	$(MKDIR_P) $(includedir)/iast $(libdir)
+	$(MKDIR_P) $(includedir)/iast $(libdir) $(pkgconfigdir)
 	$(INSTALL_DATA) $(INCLUDES) $(includedir)/iast
 	$(INSTALL_DATA) iast.h $(includedir)
+	$(INSTALL_DATA) $(PROGNAME).pc $(pkgconfigdir)
 	cp -P $(ANAME) $(LINKNAME) $(libdir)
 
 uninstall: uninstall-prog uninstall-lib uninstall-dev
@@ -108,10 +110,11 @@ uninstall-lib:
 
 uninstall-dev:
 	$(RM) -r $(includedir)/iast $(includedir)/iast.h
+	$(RM) $(pkgconfigdir)/$(PROGNAME).pc
 
 clean:
 	$(RM) $(PROGNAME) tests/test $(PROGNAME).1.gz
-	$(RM) -r $(LIBFILES) *.o */*.o */*.d *.d
+	$(RM) -r $(LIBFILES) *.o */*.o */*.d *.d $(PROGNAME).pc
 	$(RM) *.gcda *.gcno tests/coverage */*.test
 
 clean-aux:

README

@@ -16,6 +16,12 @@ For more details on the usage of the program and the requirements for the input
 data, see the included manual page iast(1).
 
 
+Donations
+---------
+If you like this project and it is useful to you, consider sending a donation
+to the following Bitcoin address: 1LXAkkvKodKB237yayzCTYAsb8tYawGfHz
+
+
 Licensing Notice
 ----------------
 This software is released under the terms of the GPL license version 2 as

compat.h

@@ -8,6 +8,7 @@
 #include <fcntl.h>
 #include <stdarg.h>
 #include <getopt.h>
+#include <ctype.h>
 #include <errno.h>
 
 #define ARRAY_SIZE(a) sizeof(a) / sizeof(*a)

configure.ac

@@ -2,7 +2,7 @@
 # Process this file with autoconf to produce a configure script.
 
 AC_PREREQ([2.69])
-AC_INIT([sanskrit-iast], [3.0.0], [vlastavesely@protonmail.ch])
+AC_INIT([sanskrit-iast], [3.1.0], [vlastavesely@protonmail.ch])
 AC_CONFIG_SRCDIR([transliteration.c])
 AC_CONFIG_HEADERS([config.h])
 
@@ -17,6 +17,9 @@ AC_PROG_INSTALL
 AC_PROG_LN_S
 AC_CHECK_TOOL([AR], [ar])
 
+PKG_PROG_PKG_CONFIG
+PKG_INSTALLDIR
+
 # Checks for libraries.
 PKG_CHECK_MODULES(CHECK, check)
 
@@ -56,5 +59,5 @@ AC_SUBST(COVERAGE_CFLAGS)
 AC_SUBST(COVERAGE_LFLAGS)
 AC_SUBST(USE_GCOV)
 
-AC_CONFIG_FILES([Makefile])
+AC_CONFIG_FILES([Makefile iast.pc])
 AC_OUTPUT

harvard-kyoto.h

@@ -3,6 +3,14 @@
 #ifndef __HARVARD_KYOTO_H
 #define __HARVARD_KYOTO_H
 
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
 int encode_harvard_kyoto_to_iast(const char *text, char **out);
 
+#if defined (__cplusplus)
+}
+#endif
+
 #endif /* __HARVARD_KYOTO_H */

iast.1

@@ -1,4 +1,4 @@
-.TH "iast" "1" "10 January 2022" "sanskrit-iast" "Sanskrit Transliteration"
+.TH "iast" "1" "02 March 2023" "sanskrit-iast" "Sanskrit Transliteration"
 
 .SH NAME
 .B iast
@@ -98,6 +98,19 @@ representation using the
 scheme.
 .RE
 
+.BR \-k ,
+.BR \-\-harvard ,
+.BR \-\-kyoto
+.RS 4
+Convert a plain
+.BR ASCII (7)
+string to the
+.I IAST
+representation using the
+.I Harvard-Kyoto
+scheme.
+.RE
+
 .BR \-a ,
 .B \-\-ascii
 .RS 4
@@ -201,14 +214,17 @@ Wikipedia:
 .br
 .I https://en.wikipedia.org/wiki/Velthuis
 - the Velthuis scheme.
+.br
+.I https://en.wikipedia.org/wiki/Harvard-Kyoto
+- the Harvard-Kyoto scheme.
 .RE
 
 
 .SH REPORTING BUGS
 If you encounter a bug, you should make sure that you are using the latest
 version of the software. If you are and the bug is still present, you can
-report it on the GitLab issues tracker:
-.IR https://gitlab.com/vlastavesely/sanskrit-iast/issues .
+report it on the GitHub issues tracker:
+.IR https://github.com/vlastavesely/sanskrit-iast/issues .
 
 
 .SH LICENSE AND WARRANTY

iast.pc.in

@@ -0,0 +1,11 @@
+prefix=@prefix@
+exec_prefix=@exec_prefix@
+libdir=@libdir@
+includedir=@includedir@
+
+Name: libiast
+URL: https://github.com/vlastavesely/sanskrit-iast
+Description: Library for romanisation of texts written in Devanagari
+Version: @PACKAGE_VERSION@
+Libs: -L${libdir} -liast
+Cflags:

tests/integration.c

@@ -152,7 +152,7 @@ END_TEST
 
 START_TEST(test_version)
 {
-	test_output("./iast -v", "iast v3.0.0\n");
+	test_output("./iast -v", "iast v3.1.0\n");
 }
 END_TEST
 

tests/translit.c

@@ -23,6 +23,18 @@ static void test_translit(const char *devanagari, const char *latin)
 	free(b);
 }
 
+static void test_translit_latin(const char *latin, const char *devanagari)
+{
+	char *str;
+	int ret;
+
+	ret = transliterate_latin_to_devanagari(latin, &str);
+	ck_assert_int_eq(0, ret);
+	ck_assert_str_eq(devanagari, str);
+
+	free(str);
+}
+
 START_TEST(test_translit_words)
 {
 	/* https://en.wikipedia.org/wiki/Sanskrit */
@@ -52,6 +64,9 @@ START_TEST(test_translit_vedic)
 
 	/* vedic stresses */
 	test_translit("अ॒ग्निमी॑ळे पु॒रोहि॑तं", "a\\_gnimī\\'ḷe pu\\_rohi\\'taṃ");
+
+	/* double svarita */
+	test_translit("अ॒ग्निमी᳚ळे", "a\\_gnimī\\\"ḷe");
 }
 END_TEST
 
@@ -67,6 +82,17 @@ START_TEST(test_translit_lla_sylable)
 }
 END_TEST
 
+START_TEST(test_translit_aum)
+{
+	test_translit("ॐ औम औमे तौमे ॐ", "aum auma aume taume aum");
+}
+
+START_TEST(test_translit_avagraha)
+{
+	test_translit_latin("śivo'ham", "शिवोऽहम्");
+	test_translit_latin("śivo’ham", "शिवोऽहम्");
+}
+
 START_TEST(test_translit_candrabindu)
 {
 	test_translit("तान्यजत्राँ", "tānyajatrām̐");
@@ -84,6 +110,8 @@ void register_translit_tests(TCase *test_case)
 	tcase_add_test(test_case, test_translit_words);
 	tcase_add_test(test_case, test_translit_vedic);
 	tcase_add_test(test_case, test_translit_lla_sylable);
+	tcase_add_test(test_case, test_translit_aum);
+	tcase_add_test(test_case, test_translit_avagraha);
 	tcase_add_test(test_case, test_translit_candrabindu);
 	tcase_add_test(test_case, test_translit_zero_width_joiner);
 }

transliteration.c

@@ -8,6 +8,7 @@
 #define SCHWA_CHARACTER   'a'
 #define ZERO_WIDTH_JOINER 0x200d
 #define VIRAMA            0x094d
+#define AUM               0x0950
 #define CHUNKSIZE         1024
 
 static struct translit_letter table[] = {
@@ -35,6 +36,7 @@ static struct translit_letter table[] = {
 	{0x0902, CODA, "\u1e43"},        /* anusvara (.m) */
 	{0x0903, CODA, "\u1e25"},        /* visarga (.h) */
 	{0x093d, CODA, "'"},             /* avagrada (') */
+	{0x093d, CODA, "’"},             /* avagrada (’) - typographic */
 	{0x0901, CODA, "m\u0310"},       /* candrabindu */
 
 	/* Consonants */
@@ -105,6 +107,7 @@ static struct translit_letter table[] = {
 	{0x0964, CODA, "|"},             /* danda */
 
 	{0x0951, CODA, "\\'"},           /* svarita */
+	{0x1cda, CODA, "\\\""},          /* double svarita */
 	{0x0952, CODA, "\\_"}            /* anudatta */
 };
 
@@ -259,6 +262,14 @@ int transliterate_latin_to_devanagari(const char *latin, char **ret)
 					letter = letter_by_code(0x0933); /* .la */
 				}
 			}
+
+			if (letter->code == AUM) {
+				/* ‘aum’ is followed by something else than
+				 * a whitespace → it is ‘au’ + ‘m…’ */
+				if (!isspace(src[3]) && src[3] != '\0') {
+					letter = letter_by_code(0x0914);
+				}
+			}
 encode_consonant:
 			/* A consonant or an initial vowel */
 			utf8_pack_char(devanagari + done, letter->code);

transliteration.h

@@ -3,6 +3,10 @@
 #ifndef __TRANSLITERATION_H
 #define __TRANSLITERATION_H
 
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
 enum translit_letter_type {
 	VOWEL,
 	CONSONANT,
@@ -30,4 +34,8 @@ static inline int is_devanagari(unsigned int code)
 	return code >= 0x0900 && code <= 0x097f;
 }
 
+#if defined (__cplusplus)
+}
+#endif
+
 #endif /* __TRANSLITERATION_H */

velthuis.h

@@ -3,7 +3,15 @@
 #ifndef __VELTHUIS_H
 #define __VELTHUIS_H
 
+#if defined (__cplusplus)
+extern "C" {
+#endif
+
 int encode_velthuis_to_iast(const char *text, char **out);
 int encode_iast_to_velthuis(const char *text, char **out);
 
+#if defined (__cplusplus)
+}
+#endif
+
 #endif /* __VELTHUIS_H */