commit 81dc4dbad250ff8e9433eea76c1cc15649a2f927
from: Sven M. Hallberg <pesco@khjk.org>
date: Tue Mar 28 17:44:29 2023 UTC

move parse_xrefs back next to main

Move parse_xrefs() back in its proper place as a helper to main(),
including the definition of the global variable 'infile' with the rest
of the command line arguments.

It had been moved in fbbe953f when the content processing code was
confusedly hooked into the function.

Also removes marker comments about "Start/End xref parsing". The code
between them is not exclusively concerned with xrefs and their sheer
size clashes with the rest of the coding style.

commit - 91473d5f1fa4f5e83430127f82ec336551db3f28
commit + 81dc4dbad250ff8e9433eea76c1cc15649a2f927
blob - 3b282184f21c92d3d3c6b73983f07c53cb54b3bf
blob + 1761f1414ca65700c8f8cea42ddf9ed29f4e7fe2
--- pdf.c
+++ pdf.c
@@ -4416,11 +4416,6 @@ process_page_content(struct Env *aux)
 
 
 /*
- * ********************************************************************
- * Start xref parsing
- * ********************************************************************
- */
-/*
  * decode the bytes in 'b' according to metadata in the stream dictionary 'd'
  * and parse the result with 'p'.
  */
@@ -4875,8 +4870,75 @@ kxstream(HAllocator *mm__, const HParsedToken *x, void
 	return h_sequence__m(mm__, dict_p, value_p, NULL);
 }
 
+
+/*
+ * main program
+ */
 
+#include <stdio.h>
+#include <string.h>	/* strtok() */
+#include <inttypes.h>
+#include <fcntl.h>	/* open() */
+#include <unistd.h>	/* lseek(), getopt() */
+#include <sys/mman.h>	/* mmap() */
+#include <errno.h>
 
+/* used with getopt() */
+extern char *optarg;
+extern int optind;
+
+/* command line arguments */
+const char *progname;
+const char *infile;
+const char *xfile, *Xfile;
+int qflag, vflag, dflag;
+int onum = -1, ogen = -1;	/* -1 means unspecified */
+
+/* helper: parse 's' as a natural number. returns -1 on error. */
+int
+strtonat(const char *s)
+{
+	char *p;
+	long l;
+
+	errno = 0;
+	l = strtol(s, &p, 10);
+	if (errno != 0)
+		return -1;
+	if (p == s || *p != '\0')	/* no parse */
+		return -1;
+	if (l < 0 || l > INT_MAX)	/* out of range */
+		return -1;
+	return l;
+}
+
+void
+usage(void)
+{
+	fprintf(stderr, "usage: %s [-qsv] [-d what] [-x out.txt] file [oid]\n",
+	    progname);
+	exit(2);
+}
+
+void
+dumpstream(FILE *f, const HParsedToken *obj)
+{
+	HParseResult *res;
+	HBytes data;
+
+	// XXX properly verify that obj is a stream (needs custom token type)
+	if (obj->token_type != TT_SEQUENCE || obj->seq->used != 2 ||
+	    obj->seq->elements[1]->token_type != TT_HParseResult)
+		errx(2, "%s: requested object is not a stream", infile);
+
+	res = H_INDEX(HParseResult, obj, 1);
+	assert(res != NULL);
+	assert(res->ast != NULL);
+	data = H_CAST_BYTES(res->ast);
+
+	fwrite(data.token, 1, data.len, f);
+}
+
 /*
  * This helper implements the standard backwards parsing strategy to read all
  * cross-reference sections and trailer dictionaries, starting from the
@@ -4887,8 +4949,6 @@ kxstream(HAllocator *mm__, const HParsedToken *x, void
  * number of elements.
  */
 // XXX review changes to this function. see git -L /^parse_xrefs/,/^}/:pdf.c'
-const char *infile = NULL;
-
 void
 parse_xrefs(struct Env *aux)
 {
@@ -4979,84 +5039,8 @@ parse_xrefs(struct Env *aux)
 end:
 	aux->xrefs = xrefs;
 	aux->nxrefs = n;
-}
-
-/*
- * ********************************************************************
- * End xref parsing
- * ********************************************************************
- */
-
-
-
-
-/*
- * main program
- */
-
-#include <stdio.h>
-#include <string.h>	/* strtok() */
-#include <inttypes.h>
-#include <fcntl.h>	/* open() */
-#include <unistd.h>	/* lseek(), getopt() */
-#include <sys/mman.h>	/* mmap() */
-#include <errno.h>
-
-/* used with getopt() */
-extern char *optarg;
-extern int optind;
-
-/* command line arguments */
-const char *progname;
-const char *xfile, *Xfile;
-int qflag, vflag, dflag;
-int onum = -1, ogen = -1;	/* -1 means unspecified */
-
-/* helper: parse 's' as a natural number. returns -1 on error. */
-int
-strtonat(const char *s)
-{
-	char *p;
-	long l;
-
-	errno = 0;
-	l = strtol(s, &p, 10);
-	if (errno != 0)
-		return -1;
-	if (p == s || *p != '\0')	/* no parse */
-		return -1;
-	if (l < 0 || l > INT_MAX)	/* out of range */
-		return -1;
-	return l;
 }
 
-void
-usage(void)
-{
-	fprintf(stderr, "usage: %s [-qsv] [-d what] [-x out.txt] file [oid]\n",
-	    progname);
-	exit(2);
-}
-
-void
-dumpstream(FILE *f, const HParsedToken *obj)
-{
-	HParseResult *res;
-	HBytes data;
-
-	// XXX properly verify that obj is a stream (needs custom token type)
-	if (obj->token_type != TT_SEQUENCE || obj->seq->used != 2 ||
-	    obj->seq->elements[1]->token_type != TT_HParseResult)
-		errx(2, "%s: requested object is not a stream", infile);
-
-	res = H_INDEX(HParseResult, obj, 1);
-	assert(res != NULL);
-	assert(res->ast != NULL);
-	data = H_CAST_BYTES(res->ast);
-
-	fwrite(data.token, 1, data.len, f);
-}
-
 int
 main(int argc, char *argv[])
 {