Unix & Linux Stack Exchange is a question and answer site for users of Linux, FreeBSD and other Un*x-like operating systems. It only takes a minute to sign up.
Sign up to join this community
If I create a PDF from LibreOffice Writer using the cups-pdf printer on Ubuntu 16.04, the resulting document displays fine, but I cannot copy/paste from it because I obtain garbled text.
The problem is restricted to cups-pdf because if I print the same document with LO Writer's "Export to PDF" button, copy/paste works well.
Is there any special setting (e.g. encoding) that must be configured on the cups-pdf printer to fix this?
I confirm exactly same behaviour, and this is long lasting "bug" [1] [2], while author of 'cups-pdf' does not consider it is a bug, so existing patches are not going to be used upstream.
For Ubuntu, using mentioned PPAs should work, but i can't test it currently.
More skills need if someone use raw Slackware. For Slackware 14.2 and CUPS-PDF 3.0.1 [4], the patch [5][6] from [7] will not work, it is too old. During blind trial-and-errors, i try to prepare working one [8].
Note, you should install it manually, then very carefully install your new 'cups-pdf' file-by-file; i can't get to work provided SlackBuild and install.sh in any way, too few skills i have. But the fact that finally problem was gone.
You should remove your existing PDF printers, and add one using manually provided CUPS-PDF_opt.ppd, optionally (to ensure) remove CUPS-PDF.ppd.
[2] https://github.com/alexivkin/CUPS-PDF-to-PDF
[3] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=658004
[4] https://slackbuilds.org/repository/14.2/office/cups-pdf/
[6] https://launchpadlibrarian.net/153781216/04_add_pdf_passthrough_support.patch
[7] https://bugs.launchpad.net/ubuntu/+source/cups-pdf/+bug/820820
[8]
--- a/extra/CUPS-PDF_noopt.ppd
+++ b/extra/CUPS-PDF_noopt.ppd
@@ -31,7 +31,8 @@
*ModelName: "Generic CUPS-PDF Printer"
*ShortNickName: "Generic CUPS-PDF Printer"
*NickName: "Generic CUPS-PDF Printer (no options)"
-*1284DeviceID: "MFG:Generic;MDL:CUPS-PDF Printer;DES:Generic CUPS-PDF Printer;CLS:PRINTER;CMD:POSTSCRIPT;"
+*1284DeviceID: "MFG:Generic;MDL:CUPS-PDF Printer;DES:Generic CUPS-PDF Printer;CLS:PRINTER;CMD:PDF,POSTSCRIPT;"
+*cupsFilter: "application/pdf 0 -"
*% cupsFilter: "application/vnd.cups-postscript 0 pstitleiconv"
*PSVersion: "(2017.000) 0"
*LanguageLevel: "2"
--- a/extra/CUPS-PDF_opt.ppd
+++ b/extra/CUPS-PDF_opt.ppd
@@ -31,7 +31,8 @@
*ModelName: "Generic CUPS-PDF Printer"
*ShortNickName: "Generic CUPS-PDF Printer"
*NickName: "Generic CUPS-PDF Printer (w/ options)"
-*1284DeviceID: "MFG:Generic;MDL:CUPS-PDF Printer;DES:Generic CUPS-PDF Printer;CLS:PRINTER;CMD:POSTSCRIPT;"
+*1284DeviceID: "MFG:Generic;MDL:CUPS-PDF Printer;DES:Generic CUPS-PDF Printer;CLS:PRINTER;CMD:PDF,POSTSCRIPT;"
+*cupsFilter: "application/pdf 0 -"
*% cupsFilter: "application/vnd.cups-postscript 0 pstitleiconv"
*PSVersion: "(2017.000) 0"
*LanguageLevel: "2"
--- a/src/cups-pdf.c
+++ b/src/cups-pdf.c
@@ -60,6 +60,7 @@
extern int errno;
static FILE *logfp=NULL;
+int input_is_pdf=0;
static void log_event(short type, char message[], char *detail) {
time_t secs;
@@ -427,7 +427,7 @@
int len;
cp_string setup;
- printf("file cups-pdf:/ \"Virtual PDF Printer\" \"CUPS-PDF\" \"MFG:Generic;MDL:CUPS-PDF Printer;DES:Generic CUPS-PDF Printer;CLS:PRINTER;CMD:POSTSCRIPT;\"\n");
+ printf("file cups-pdf:/ \"Virtual PDF Printer\" \"CUPS-PDF\" \"MFG:Generic;MDL:CUPS-PDF Printer;DES:Generic CUPS-PDF Printer;CLS:PRINTER;CMD:PDF,POSTSCRIPT;\"\n");
if ((dir = opendir(CP_CONFIG_PATH)) != NULL) {
while ((config_ent = readdir(dir)) != NULL) {
@@ -460,6 +461,7 @@
cp_string buffer;
int rec_depth,is_title=0;
FILE *fpdest;
+ size_t bytes = 0;
if (fpsrc == NULL) {
log_event(CPERROR, "failed to open source stream", NULL);
@@ -483,14 +485,39 @@
log_event(CPSTATUS, "***Experimental Option: FixNewlines");
else
log_event(CPDEBUG, "using traditional fgets");
- while (fgets2(buffer, BUFSIZE, fpsrc) != NULL) {
+ while ((bytes = fread(buffer, sizeof(char), 4, fpsrc)) > 0) {
+ if (!strncmp(buffer, "%PDF", 4)) {
+ log_event(CPDEBUG, "found beginning of PDF code", buffer);
+ input_is_pdf=1;
+ break;
+ }
if (!strncmp(buffer, "%!", 2) && strncmp(buffer, "%!PS-AdobeFont", 14)) {
log_event(CPDEBUG, "found beginning of postscript code: %s", buffer);
break;
}
}
- log_event(CPDEBUG, "now extracting postscript code");
- (void) fputs(buffer, fpdest);
+ log_event(CPDEBUG, "now extracting code");
+ fwrite(buffer, sizeof(char), bytes, fpdest);
+ if (input_is_pdf) {
+ while((bytes = fread(buffer, sizeof(char), BUFSIZE, fpsrc)) > 0)
+ fwrite(buffer, sizeof(char), bytes, fpdest);
+/* Commented out because decoding of utf16 PDF strings isn't implemented.
+ rewind(fpsrc);
+ while (fgets2(buffer, BUFSIZE, fpsrc) != NULL) {
+ if (!is_title) {
+ char *begin = strstr(buffer, "/Title");
+ if (begin) {
+ char *end = strstr(begin, ">");
+ if (end) {
+ strncpy(title, begin+6, BUFSIZE);
+ log_event(CPDEBUG, "found title in PDF code", title);
+ is_title=1;
+ }
+ }
+ }
+ }*/
+ }
+ else {
while (fgets2(buffer, BUFSIZE, fpsrc) != NULL) {
(void) fputs(buffer, fpdest);
if (!is_title && !rec_depth)
@@ -513,6 +540,7 @@
}
}
}
+ }
(void) fclose(fpdest);
(void) fclose(fpsrc);
log_event(CPDEBUG, "all data written to spoolfile: %s", spoolfile);
@@ -761,7 +789,12 @@
(void) fclose(logfp);
return 5;
}
- snprintf(gscall, size, Conf_GSCall, Conf_GhostScript, Conf_PDFVer, outfile, spoolfile);
+ if (input_is_pdf) {
+ snprintf(gscall, size, "cp %s %s", spoolfile, outfile);
+ }
+ else {
+ snprintf(gscall, size, Conf_GSCall, Conf_GhostScript, Conf_PDFVer, outfile, spoolfile);
+ }
log_event(CPDEBUG, "ghostscript commandline built: %s", gscall);
(void) unlink(outfile);
Well you can also try this method. How about using the PDF export feature in Libreoffice. You can then print or cups-pdf or whatever you want from that point on.
