aleleba/pdfcraft
1
0
Fork 0
mirror of https://github.com/storytold/pdfcraft.git synced 2026-11-03 06:51:50 -06:00
Code Issues Packages Projects Releases Wiki Activity

Edit text: do not group or join by an empty font name (#3661)

Browse Source 
Review follow-up. `Metrics::from_dict` leaves `base_font` empty when a font
has no `/BaseFont`, and a bare subset tag such as `/ABCDEF+` left an empty
name after stripping. Two consequences:

- `same_style` split one paragraph into separate reading blocks for the
  exporters when both names were empty, though they grouped before subset
  joining;
- `/ABCDEF+` and `/GHIJKL+` compared equal after stripping, so unrelated
  malformed fonts could join into one line.

`same_base_font` now strips both tags first and requires a non-empty, equal
name, so a missing name or a bare tag matches nothing. `same_style` keeps
grouping lines whose fonts both lack a name, since the resource or fallback
face `same_font` already matched is their whole identity.

Tests: a page in a font without `/BaseFont` still reads as one block (it
split into two before this change), and a bare tag matches no other name.
This commit is contained in:
Aman Mehtar
2026-10-20 22:28:08 +05:30
parent 5eae8d6e90
commit 4a618e8dcb
3 changed files with 38 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
crates/edit/src/tests.rs
+19 -1
View File
@@ -883,7 +883,7 @@ fn text_page(content: &str) -> Document {
let objs: Vec<String> = vec![
"<< /Type /Catalog /Pages 2 0 R >>".into(),
"<< /Type /Pages /Kids [3 0 R] /Count 1 >>".into(),
"<< /Type /Page /Parent 2 0 R /MediaBox [0 0 600 800] /Contents 4 0 R /Resources << /Font << /F1 5 0 R /F2 6 0 R /F3 7 0 R /F4 8 0 R >> >> >>".into(),
"<< /Type /Page /Parent 2 0 R /MediaBox [0 0 600 800] /Contents 4 0 R /Resources << /Font << /F1 5 0 R /F2 6 0 R /F3 7 0 R /F4 8 0 R /F5 9 0 R >> >> >>".into(),
format!("<< /Length {} >>\nstream\n{content}\nendstream", content.len()),
"<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica /Encoding /WinAnsiEncoding >>".into(),
// Subset: glyphs for a (97) and b (98) only.
@@ -894,6 +894,8 @@ fn text_page(content: &str) -> Document {
.into(),
// Helvetica again, in a second resource.
"<< /Type /Font /Subtype /Type1 /BaseFont /Helvetica /Encoding /WinAnsiEncoding >>".into(),
// No `/BaseFont`: `Metrics::from_dict` leaves its base-font name empty.
"<< /Type /Font /Subtype /Type1 /Encoding /WinAnsiEncoding >>".into(),
];
let mut out = b"%PDF-1.7\n".to_vec();
let mut offs = Vec::new();
@@ -1760,6 +1762,22 @@ fn different_fonts_start_a_new_line() {
assert_eq!(text::text_lines(&doc, 0).unwrap().len(), 2);
}
/// A font without `/BaseFont` has no name to compare. Its lines still group by resource, as they
/// did before subset joining: an empty name must not split one paragraph into two blocks.
#[test]
fn lines_in_a_font_without_a_base_font_stay_one_block() {
let doc = text_page("BT /F5 12 Tf 72 700 Td (First) Tj 0 -15 Td (second) Tj ET");
let lines = text::text_lines(&doc, 0).unwrap();
assert_eq!(lines.len(), 2);
assert!(lines.iter().all(|l| l.base_font.is_empty()), "{:?}", lines.iter().map(|l| &l.base_font).collect::<Vec<_>>());
let blocks = text::text_blocks(&doc, 0).unwrap();
assert_eq!(blocks.len(), 1);
assert_eq!(blocks[0].text, "First second");
let reading = text::reading_blocks(&doc, 0).unwrap();
assert_eq!(reading.len(), 1, "{:?}", reading.iter().map(|b| &b.text).collect::<Vec<_>>());
assert_eq!(reading[0].text, "First second");
}
#[test]
fn text_lines_are_found_and_replaced_in_place() {
let mut doc =