From 76e546cef80d23e018a051db4b5b29ed1f409b20 Mon Sep 17 00:00:00 2001
From: "Sven M. Hallberg" <pesco@khjk.org>
Date: Tue, 14 Mar 2023 17:02:45 +0000
Subject: [PATCH] process page content after main parser and only for text
extraction
Factors this code out of parse_xrefs() where it never belonged, into a
new function process_page_content() that is called from main after the
main parse has succeeded and only if text extraction was requested, i.e.
-x or -X was passed on the command line.
Also adjusts the code for style and drops some related XXXs.
Fixes #49.
---
pdf.c | 118 ++++++++++++++++++++++++++++++----------------------------
1 file changed, 61 insertions(+), 57 deletions(-)
diff --git a/pdf.c b/pdf.c
index 49fc63a..72d1ed3 100644
--- a/pdf.c
+++ b/pdf.c
@@ -4356,6 +4356,60 @@ parse_catalog(struct Env *aux, const HParsedToken *root)
}
}
+
+void
+process_page_content(struct Env *aux)
+{
+ const HParsedToken *tok, *xref;
+ Dict *trailer;
+ size_t Size;
+
+ /* use the last xref section */
+ if (aux->nxrefs == 0)
+ return;
+ xref = aux->xrefs[aux->nxrefs - 1];
+
+ /* make sure we parsed a valid trailer dictionary */
+ tok = H_INDEX_TOKEN(xref, 1);
+ trailer = H_CAST(Dict, tok);
+ if (trailer == NULL) {
+ // XXX this cannot actually happen (rule xr_td)
+ log_message(7, "VIOLATION[7]: Invalid Trailer Section "
+ "or Trailer Section not found\n");
+ return;
+ }
+
+ /* fetch Size (a required field) from the trailer dictionary */
+ tok = dictentry(trailer, "Size");
+ if (tok == NULL || tok->token_type != TT_SINT) {
+ log_message(5, "VIOLATION[5]: error parsing trailer section!"
+ "Missing or malformed /Size field.\n");
+ return;
+ }
+ Size = H_CAST_SINT(tok);
+
+ // XXX this likely belongs somewhere else
+ // and it is wrong, since nxrefs is the number of xref _sections_, not
+ // the number of xref entries (objects) as this code assumes. m(
+#if 0
+ if (aux->nxrefs > Size) {
+ log_message(7, "%s: Number of xrefs found (%zu) "
+ "Greater than specified /Size (%zu).\n"
+ "Ignoring objects numbered greater than (%zu)!\n",
+ infile, aux->nxrefs, Size, n);
+ aux->nxrefs = Size;
+ }
+#else
+ (void)Size; // XXX silence warning
+#endif
+
+ /* process the document starting from the /Root object */
+ tok = dictentry(trailer, "Root");
+ if (tok != NULL && tok->token_type == TT_Ref)
+ parse_catalog(aux, tok);
+}
+
+
/*
* ********************************************************************
* End Catalog parsing
@@ -4853,11 +4907,9 @@ parse_xrefs(struct Env *aux)
const HParsedToken *tok = NULL;
size_t n = 0, nfwd = 0;
size_t offset = 0;
- bool processRoot = true;
- size_t maxObjNum = 0;
- Dict *trailer = NULL;
// Initialize the environment variables
+ // XXX not needed?!
aux->xrefs = xrefs;
aux->nxrefs = n;
@@ -4907,25 +4959,6 @@ parse_xrefs(struct Env *aux)
err(2, "parse_xrefs");
xrefs[n++] = res->ast;
-
- /* process the root */
- // XXX this validation likely belongs someplace else
- if (processRoot) {
- // Size is a required field in the trailer dictionary
- const HParsedToken *trailer_t = H_INDEX_TOKEN(res->ast, 1);
- trailer = H_CAST(Dict, trailer_t);
- const HParsedToken *size_t = dictentry(trailer, "Size");
- if (size_t == NULL || size_t->token_type != TT_SINT) {
- log_message(5, "VIOLATION[5]: error parsing trailer section!"
- "Missing or malformed -Size- field\n");
- return;
- }
- maxObjNum = H_CAST_SINT(size_t);
-
- processRoot = false;
- }
-
-
/* look up the next offset (to the previous xref section) */
tok = dictentry(H_INDEX(Dict, res->ast, 1), "Prev");
if (tok == NULL)
@@ -4958,42 +4991,10 @@ parse_xrefs(struct Env *aux)
offset = (size_t)tok->sint;
}
-
-
- /* Make sure we parsed a valid trailer section */
- if (! trailer) {
- log_message(7, "VIOLATION[7]: Invalid Trailer Section or Trailer Section not found\n");
- return;
- }
-
-
-
aux->xrefs = xrefs;
aux->nxrefs = n;
-
-
-
- // XXX likely belongs somewhere else (see XXX in the loop above)
- if (n > maxObjNum) {
- log_message(7, "%s: Number of xrefs found -%ld- "
- "Greater than specified /Size -%ld-.\n"
- "Ignoring objects numbered greater than -%ld-!\n",
- infile, n, maxObjNum, n);
- aux->nxrefs = maxObjNum;
- }
-
-
- // Process the trailer dictionary
- // XXX content processing does not belong in this function
- const HParsedToken *root = dictentry(trailer, "Root");
- if(root != NULL && root->token_type == TT_Ref)
- parse_catalog(aux, root);
-
- return;
}
-
-
/*
* ********************************************************************
* End xref parsing
@@ -5210,9 +5211,12 @@ main(int argc, char *argv[])
h_pprintln(stdout, res->ast);
}
- /* Save the extracted text */
- if (aux.ntextobjs > 0)
- text_extract(&aux, xfile, Xfile);
+ /* extract text if requested */
+ if (xfile != NULL || Xfile != NULL) {
+ process_page_content(&aux);
+ if (aux.ntextobjs > 0) // XXX always create output file?
+ text_extract(&aux, xfile, Xfile);
+ }
return 0;
}
--
GitLab