Skip to content

pdftopdf: don't abort on annotations that have no appearance stream #255

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tillkamppeter merged 1 commit into from
Sep 28, 2026
Merged

pdftopdf: don't abort on annotations that have no appearance stream #255

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions Makefile.am
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -96,7 +96,8 @@ check_SCRIPTS = \
cupsfilters/test-pclm-overflow.sh \
cupsfilters/test-pwgtopdf-bit-row.sh \
cupsfilters/test-pdftoraster-copy-height.sh \
cupsfilters/test-pdftopdf-inherited-mediabox.sh
cupsfilters/test-pdftopdf-inherited-mediabox.sh \
cupsfilters/test-pdftopdf-no-appearance.sh

check_PROGRAMS = \
testcmyk \
Expand Down Expand Up @@ -124,7 +125,8 @@ TESTS = \
cupsfilters/test-pclm-overflow.sh \
cupsfilters/test-pwgtopdf-bit-row.sh \
cupsfilters/test-pdftoraster-copy-height.sh \
cupsfilters/test-pdftopdf-inherited-mediabox.sh
cupsfilters/test-pdftopdf-inherited-mediabox.sh \
cupsfilters/test-pdftopdf-no-appearance.sh

# testcmyk # fails as it opens some image.ppm which is nowerhe to be found.
# testimage # requires also some ppm file as argument
Expand Down Expand Up @@ -361,6 +363,7 @@ EXTRA_DIST += cupsfilters/gen-lorem-text.c
# ships in "make dist".
EXTRA_DIST += cupsfilters/test-pdftoraster-copy-height.c
EXTRA_DIST += cupsfilters/test-pdftopdf-inherited-mediabox.c
EXTRA_DIST += cupsfilters/test-pdftopdf-no-appearance.c
# Generated deterministic lorem text for texttopdf tests
BUILT_SOURCES = cupsfilters/test_files/test_text_lorem.txt
CLEANFILES = cupsfilters/test_files/test_text_lorem.txt
Expand All @@ -379,6 +382,7 @@ EXTRA_DIST += \
cupsfilters/test_files/filled-2.pdf \
cupsfilters/test_files/form_english.pdf \
cupsfilters/test_files/inherited_mediabox.pdf \
cupsfilters/test_files/link-annots-no-appearance.pdf \
cupsfilters/test_files/malformed.pdf \
cupsfilters/test_files/onepage-a4-adobe-rgb-8-150dpi.pwg \
cupsfilters/test_files/test_file_1pg.pdf \
Expand Down
257 changes: 143 additions & 114 deletions cupsfilters/pdftopdf.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1595,7 +1595,7 @@ flatten_pdf(xform_prepare_t *p, // I - Preparation data
rotate_val = page_get_rotate(outpage->input[pg]);
count = pdfioArrayGetSize(annotsArray);

p->annotation_contents = (char**)malloc(count * sizeof(char*));
p->annotation_contents = (char**)calloc(count, sizeof(char*));

int* noAppearanceobjectIndex = (int *)malloc(count * sizeof(int));
noAppearanceobjectCount = 0;
Expand Down Expand Up @@ -1840,148 +1840,177 @@ flatten_pdf(xform_prepare_t *p, // I - Preparation data
fprintf(stderr, "ignore annotation with no appearance\n");
noAppearanceobjectCount++;
}
else
{
char *name = (char *)malloc(sizeof(char) * 32);
snprintf(name, 32, "Fxo%d", next_fx);

pdfio_dict_t *page_resources = pdfioDictGetDict(outpage->pagedict, "Resources");
if (!page_resources)
else
{
// No selected appearance stream (/N). Link annotations and form
// fields with no appearance have nothing to paint. This branch used
// to build a Form XObject anyway and register it under a
// heap-allocated name, then free that name. pdfioDictSetObj() keeps
// the caller's key pointer, so by the time the flattened file was
// written every /XObject key was a dangling pointer. Reopening the
// file failed and pdftopdf exited 1. Issue #246.
//
// Text and choice fields that still carry a value are synthesized
// below. The XObject name is a PDFio-owned string.
const char *field_type,
*field_value = NULL,
*da_string = NULL;

field_type = pdfioDictGetName(Annot_dict, "FT");
if (field_type && (strcmp(field_type, "Tx") == 0 ||
strcmp(field_type, "Ch") == 0))
{
page_resources = pdfioDictCreate(outpage->pdf);
pdfioDictSetDict(outpage->pagedict, "Resources", page_resources);
}
field_value = pdfioDictGetName(Annot_dict, "V");
if (!field_value)
field_value = pdfioDictGetString(Annot_dict, "V");

pdfio_dict_t *xobj_dict = pdfioDictGetDict(page_resources, "XObject");
if (!xobj_dict)
{
xobj_dict = pdfioDictCreate(outpage->pdf);
pdfioDictSetDict(page_resources, "XObject", xobj_dict);
da_string = pdfioDictGetName(Annot_dict, "DA");
if (!da_string)
da_string = pdfioDictGetString(Annot_dict, "DA");
}

pdfio_array_t *procset = pdfioArrayCreate(outpage->pdf);
pdfioArrayAppendName(procset, "PDF"); // adds /PDF
pdfioArrayAppendName(procset, "Text"); // adds /Text

pdfio_dict_t *resources = pdfioDictCreate(outpage->pdf);
pdfioDictSetArray(resources, "ProcSet", procset);
if (!field_value || !da_string)
{
fprintf(stderr, "DEBUG: special case ignore annotation with no appearance\n");
noAppearanceobjectIndex[noAppearanceobjectCount] = (int)i;
noAppearanceobjectCount++;
}
else
{
const char *name;
char *content;
pdfio_dict_t *page_resources,
*xobj_dict,
*form_xobj_dict,
*resources;
pdfio_array_t *procset;
pdfio_obj_t *form_xobj;
pdfio_rect_t rect,
bbox;
char font_key[64];
double font_size = 10.0;

name = pdfioStringCreatef(outpage->pdf, "Fxo%d", next_fx);
content = name ? special_pdfio_annotation_get_content(Annot_obj, name,
rotate_val, forbidden_flags, required_flags) : NULL;
if (!name || !content || content[0] == '\0')
{
free(content);
fprintf(stderr, "DEBUG: special case ignore annotation with no appearance\n");
noAppearanceobjectIndex[noAppearanceobjectCount] = (int)i;
noAppearanceobjectCount++;
}
else
{
page_resources = pdfioDictGetDict(outpage->pagedict, "Resources");
if (!page_resources)
{
page_resources = pdfioDictCreate(outpage->pdf);
pdfioDictSetDict(outpage->pagedict, "Resources", page_resources);
}

xobj_dict = pdfioDictGetDict(page_resources, "XObject");
if (!xobj_dict)
{
xobj_dict = pdfioDictCreate(outpage->pdf);
pdfioDictSetDict(page_resources, "XObject", xobj_dict);
}

pdfio_dict_t *form_xobj_dict = pdfioDictCreate(outpage->pdf);
pdfioDictSetName(form_xobj_dict, "Type", "XObject");
pdfioDictSetName(form_xobj_dict, "Subtype", "Form");
procset = pdfioArrayCreate(outpage->pdf);
pdfioArrayAppendName(procset, "PDF");
pdfioArrayAppendName(procset, "Text");

char *content = special_pdfio_annotation_get_content(Annot_obj, name, rotate_val, forbidden_flags, required_flags);
p->annotation_contents[i-noAppearanceobjectCount] = content;
resources = pdfioDictCreate(outpage->pdf);
pdfioDictSetArray(resources, "ProcSet", procset);

// The addition to page xobject should be made only if the content stream is not NULL.
if (content && content[0] != '\0')
{
pdfio_array_t *bbox = pdfioDictGetArray(Annot_dict, "BBox");
if (bbox)
{
pdfioDictSetArray(form_xobj_dict, "BBox", pdfioArrayCopy(outpage->pdf, bbox));
}
else
{
fprintf(stderr, "WARNING: Appearance stream is missing required /BBox.\n");
pdfio_rect_t rect;

if (pdfioDictGetRect(Annot_dict, "Rect", &rect))
{
pdfio_rect_t Bbox;
Bbox.x1 = 0;
Bbox.y1 = 0;
Bbox.x2 = rect.x2 - rect.x1;
Bbox.y2 = rect.y2 - rect.y1;
pdfioDictSetRect(form_xobj_dict, "BBox", &Bbox);
}
}

pdfio_obj_t *form_xobj = pdfioFileCreateObj(outpage->pdf, form_xobj_dict);

const char *field_type = pdfioDictGetName(Annot_dict, "FT");
if (field_type && (strcmp(field_type, "Tx") == 0 || strcmp(field_type, "Ch") == 0))
{
const char *field_value = pdfioDictGetName(Annot_dict, "V");
if (!field_value) field_value = pdfioDictGetString(Annot_dict, "V");
form_xobj_dict = pdfioDictCreate(outpage->pdf);
pdfioDictSetName(form_xobj_dict, "Type", "XObject");
pdfioDictSetName(form_xobj_dict, "Subtype", "Form");

if (!pdfioDictGetRect(Annot_dict, "Rect", &rect))
{
rect.x1 = 0.0;
rect.y1 = 0.0;
rect.x2 = 0.0;
rect.y2 = 0.0;
}
bbox.x1 = 0.0;
bbox.y1 = 0.0;
bbox.x2 = rect.x2 - rect.x1;
bbox.y2 = rect.y2 - rect.y1;
pdfioDictSetRect(form_xobj_dict, "BBox", &bbox);

const char *da_string = pdfioDictGetName(Annot_dict, "DA");
if (!da_string) da_string = pdfioDictGetString(Annot_dict, "DA");
form_xobj = pdfioFileCreateObj(outpage->pdf, form_xobj_dict);

char font_key[64];
double font_size = 10.0;
if (extractFontDetails(da_string, font_key, sizeof(font_key), &font_size))
{
pdfio_obj_t *font_obj = NULL;
pdfio_dict_t *page_resource_dict = pdfioDictGetDict(outpage->pagedict, "Resources");
pdfio_dict_t *font_dict = page_resource_dict ? pdfioDictGetDict(page_resource_dict, "Font") : NULL;

{
pdfio_obj_t *font_obj = NULL;
pdfio_dict_t *page_resource_dict,
*font_dict;

page_resource_dict = pdfioDictGetDict(outpage->pagedict, "Resources");
font_dict = page_resource_dict ? pdfioDictGetDict(page_resource_dict, "Font") : NULL;
if (font_dict)
font_obj = pdfioDictGetObj(font_dict, font_key);

if (!font_obj)
{
font_obj = pdfioDictGetObj(font_dict, font_key);
}

// If font_obj is not found in the page's resources, check the AcroForm's /DR.
if (!font_obj)
{
pdfio_dict_t *catalog = pdfioFileGetCatalog(p->inpdf);
pdfio_dict_t *acroform = catalog ? pdfioDictGetDict(catalog, "AcroForm") : NULL;
pdfio_dict_t *acroform_dr = acroform ? pdfioDictGetDict(acroform, "DR") : NULL;
pdfio_dict_t *acroform_font_dict = acroform_dr ? pdfioDictGetDict(acroform_dr, "Font") : NULL;
pdfio_dict_t *catalog,
*acroform,
*acroform_dr,
*acroform_font_dict;

catalog = pdfioFileGetCatalog(p->inpdf);
acroform = catalog ? pdfioDictGetDict(catalog, "AcroForm") : NULL;
acroform_dr = acroform ? pdfioDictGetDict(acroform, "DR") : NULL;
acroform_font_dict = acroform_dr ? pdfioDictGetDict(acroform_dr, "Font") : NULL;
if (acroform_font_dict)
{
font_obj = pdfioDictGetObj(acroform_font_dict, font_key);
}
font_obj = pdfioDictGetObj(acroform_font_dict, font_key);
}
if (font_obj)
{
// 1. Create the dedicated sub-dictionary for fonts.
pdfio_dict_t *sub_font_dict = pdfioDictCreate(outpage->pdf);
pdfioDictSetObj(sub_font_dict, font_key, font_obj);

pdfioDictSetObj(sub_font_dict, pdfioStringCreate(outpage->pdf, font_key), font_obj);
pdfioDictSetDict(resources, "Font", sub_font_dict);
pdfioDictSetDict(form_xobj_dict, "Resources", resources);

fprintf(stderr, "SUCCESS: Font /%s correctly nested in /Resources /Font dictionary.\\n", font_key);
}
else
{
fprintf(stderr, "ERROR: Font %s not found in page or AcroForm resources.\\n", font_key);
}
}
else
fprintf(stderr, "ERROR: Font %s not found in page or AcroForm resources.\n", font_key);
}

if (field_value && da_string)
{
if (form_xobj)
{
pdfio_stream_t *dst_stream = pdfioObjCreateStream(form_xobj, PDFIO_FILTER_NONE);

if (dst_stream)
{
pdfio_rect_t form_bbox;
if (pdfioDictGetRect(form_xobj_dict, "BBox", &form_bbox))
{
double field_height = form_bbox.y2 - form_bbox.y1;
double x = 2.0;
double y = (field_height / 2.0) - (font_size * 0.35);

pdfioStreamPuts(dst_stream, "BT\n");
pdfioStreamPrintf(dst_stream, "%s\n", da_string); // Assuming da_string is safe
pdfioStreamPrintf(dst_stream, "%.2f %.2f Td\n", x, y);
pdfioStreamPrintf(dst_stream, "(%s) Tj\n", field_value); // The critical fix
pdfioStreamPuts(dst_stream, "ET\n");
pdfio_rect_t form_bbox;
double field_height,
x,
y;

if (pdfioDictGetRect(form_xobj_dict, "BBox", &form_bbox))
{
field_height = form_bbox.y2 - form_bbox.y1;
x = 2.0;
y = (field_height / 2.0) - (font_size * 0.35);
pdfioStreamPuts(dst_stream, "BT\n");
pdfioStreamPrintf(dst_stream, "%s\n", da_string);
pdfioStreamPrintf(dst_stream, "%.2f %.2f Td\n", x, y);
pdfioStreamPrintf(dst_stream, "(%s) Tj\n", field_value);
pdfioStreamPuts(dst_stream, "ET\n");
}
pdfioStreamClose(dst_stream);
}
}
}

pdfioDictSetObj(xobj_dict, name, form_xobj);
next_fx++;
}

pdfioDictSetObj(xobj_dict, name, form_xobj);
p->annotation_contents[i - noAppearanceobjectCount] = content;
next_fx++;
}
}

//Annot_dict
free(name);
fprintf(stderr, "DEBUG: special case ignore annotation with no appearance\n");
noAppearanceobjectIndex[noAppearanceobjectCount] = i;
}
}
if(N_stream)
pdfioStreamClose(N_stream);
}
Expand Down
41 changes: 41 additions & 0 deletions cupsfilters/test-pdftopdf-no-appearance.c
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,41 @@
//
// Regression check for issue #246: a PDF whose only annotations are links
// with no appearance stream must come back out of pdftopdf as a readable
// one-page PDF.
//

#include <stdio.h>
#include <pdfio.h>

int
main(int argc,
char *argv[])
{
pdfio_file_t *pdf;
size_t pages;


if (argc != 2)
{
fprintf(stderr, "Usage: %s file.pdf\n", argv[0]);
return (1);
}

pdf = pdfioFileOpen(argv[1], NULL, NULL, NULL, NULL);
if (!pdf)
{
fprintf(stderr, "pdftopdf output could not be opened\n");
return (1);
}

pages = pdfioFileGetNumPages(pdf);
pdfioFileClose(pdf);

if (pages < 1)
{
fprintf(stderr, "pdftopdf output has no pages\n");
return (1);
}

return (0);
}
Loading
Loading