From 85ee87997e3ee4eb579084f92d109b9b78dcf9c7 Mon Sep 17 00:00:00 2001 From: Tor Andersson Date: Thu, 25 Aug 2016 12:44:13 +0200 Subject: Use U+FFFD instead of '?' for bad encodings in text extraction. --- source/pdf/pdf-op-run.c | 2 +- source/pdf/pdf-unicode.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'source/pdf') diff --git a/source/pdf/pdf-op-run.c b/source/pdf/pdf-op-run.c index 96084cdf..aad0aebf 100644 --- a/source/pdf/pdf-op-run.c +++ b/source/pdf/pdf-op-run.c @@ -902,7 +902,7 @@ pdf_show_char(fz_context *ctx, pdf_run_processor *pr, int cid) } if (ucslen == 0 || (ucslen == 1 && ucsbuf[0] == 0)) { - ucsbuf[0] = '?'; + ucsbuf[0] = 0xFFFD; ucslen = 1; } diff --git a/source/pdf/pdf-unicode.c b/source/pdf/pdf-unicode.c index ca84341d..65bda460 100644 --- a/source/pdf/pdf-unicode.c +++ b/source/pdf/pdf-unicode.c @@ -96,7 +96,7 @@ pdf_load_to_unicode(fz_context *ctx, pdf_document *doc, pdf_font_desc *font, if (strings[cpt]) font->cid_to_ucs[cpt] = pdf_lookup_agl(strings[cpt]); else - font->cid_to_ucs[cpt] = '?'; + font->cid_to_ucs[cpt] = 0xFFFD; /* replacement character */ } } -- cgit v1.2.3