fenix
/
cpdf-source
mirror of https://github.com/johnwhitington/cpdf-source.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Moves text autotagging from textsection to -text

This commit is contained in:
John Whitington 2024-10-21 13:44:20 +01:00
parent c77c9beb95
commit e39af4218a
1 changed files with 4 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
cpdfdraw.ml
View File

@ -442,16 +442,14 @@ let rec ops_of_drawop struct_tree dryrun pdf endpage filename bates batespad num
(res ()).font_size <- size; (res ()).font_size <- size;
[] []
| TextSection ops -> | TextSection ops ->
[Pdfops.Op_BT] @ ops_of_drawops struct_tree dryrun pdf endpage filename bates batespad num page ops @ [Pdfops.Op_ET]
| Text s ->
if dryrun then iter (fun c -> Hashtbl.replace (res ()).current_fontpack_codepoints c ()) (Pdftext.codepoints_of_utf8 s);
let m = mcid () in let m = mcid () in
if not dryrun && !do_auto_tag then structdata := StDataMCID ("/P", m)::!structdata; if not dryrun && !do_auto_tag then structdata := StDataMCID ("/P", m)::!structdata;
(if struct_tree && !do_auto_tag then [Pdfops.Op_BDC ("/P", Pdf.Dictionary ["/MCID", Pdf.Integer m])] else []) (if struct_tree && !do_auto_tag then [Pdfops.Op_BDC ("/P", Pdf.Dictionary ["/MCID", Pdf.Integer m])] else [])
@ [Pdfops.Op_BT] @ fst (runs_of_utf8 s)
@ ops_of_drawops struct_tree dryrun pdf endpage filename bates batespad num page ops
@ [Pdfops.Op_ET]
@ (if struct_tree && !do_auto_tag then [Pdfops.Op_EMC] else []) @ (if struct_tree && !do_auto_tag then [Pdfops.Op_EMC] else [])
| Text s ->
if dryrun then iter (fun c -> Hashtbl.replace (res ()).current_fontpack_codepoints c ()) (Pdftext.codepoints_of_utf8 s);
fst (runs_of_utf8 s)
| SpecialText s -> | SpecialText s ->
let s = process_specials pdf endpage filename bates batespad num page s in let s = process_specials pdf endpage filename bates batespad num page s in
if dryrun then iter (fun c -> Hashtbl.replace (res ()).current_fontpack_codepoints c ()) (Pdftext.codepoints_of_utf8 s); if dryrun then iter (fun c -> Hashtbl.replace (res ()).current_fontpack_codepoints c ()) (Pdftext.codepoints_of_utf8 s);