Support input files with CR/LF line terminators.

3 files changed, 37 insertions, 1 deletions

diff --git a/ChangeLog b/ChangeLog
index 75a3957..21218d7 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,9 @@
+2018-04-24  Bruno Haible  <bruno@clisp.org>
+
+	Support input files with CR/LF line terminators.
+	Reported at <https://savannah.gnu.org/bugs/?53732>.
+	* src/input.cc (Input::read_input): Convert CR/LF sequences to LF.
+
 2018-01-27  Bruno Haible  <bruno@clisp.org>
 
 	Rename some files.
diff --git a/NEWS b/NEWS
index f3abd0e..14deb90 100644
--- a/NEWS
+++ b/NEWS
@@ -1,3 +1,9 @@
+New in 3.2:
+* The input file may now use Windows line terminators (CR/LF) instead of
+  Unix line terminators (LF).
+  Note: This is an incompatible change. If you want to use a keyword that
+  ends in a CR byte, such as xyz<CR>, write it as "xyz\r".
+
 New in 3.1:
 * The generated C code is now in ANSI-C by default. If you want to support
   pre-ANSI-C compilers, you need to provide the option --language=C on the
diff --git a/src/input.cc b/src/input.cc
index 67c93b5..fca04e0 100644
--- a/src/input.cc
+++ b/src/input.cc
@@ -1,5 +1,5 @@
 /* Input routines.
-   Copyright (C) 1989-1998, 2002-2004, 2011 Free Software Foundation, Inc.
+   Copyright (C) 1989-1998, 2002-2004, 2011, 2017-2018 Free Software Foundation, Inc.
    Written by Douglas C. Schmidt <schmidt@ics.uci.edu>
    and Bruno Haible <bruno@clisp.org>.
 
@@ -263,6 +263,30 @@ Input::read_input ()
       exit (1);
     }
 
+  /* Convert CR/LF line terminators (Windows) to LF line terminators (Unix).
+     GCC 3.3 and newer support CR/LF line terminators in C sources on Unix,
+     so we do the same.
+     The so-called "text mode" in stdio on Windows translates CR/LF to \n
+     automatically, but here we also need this conversion on Unix.  As a side
+     effect, on Windows we also parse CR/CR/LF into a single \n, but this
+     is not a problem.  */
+  {
+    char *p = input;
+    char *p_end = input + input_length;
+    /* Converting the initial segment without CRs is a no-op.  */
+    while (p < p_end && *p != '\r')
+      p++;
+    /* Then start the conversion for real.  */
+    char *q = p;
+    while (p < p_end)
+      {
+        if (p[0] == '\r' && p + 1 < p_end && p[1] == '\n')
+          p++;
+        *q++ = *p++;
+      }
+    input_length = q - input;
+  }
+
   /* We use input_end as a limit, in order to cope with NUL bytes in the
      input.  But note that one trailing NUL byte has been added after
      input_end, for convenience.  */