extern/pdf-to-markdown
1
0
Fork 0
mirror of https://github.com/jzillmann/pdf-to-markdown.git synced 2024-11-21 23:33:31 +01:00
Code Issues Packages Projects Releases Wiki Activity

Prevent headline detection code from detecting headline > 6

Browse Source
This commit is contained in:
Johannes Zillmann 2017-05-02 19:26:11 +02:00
parent 908f0b4be1
commit ecde2ea0f5
1 changed files with 12 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
src/javascript/models/transformations/lineitem/DetectHeaders.jsx
View File

@ -67,16 +67,19 @@ export default class DetectHeaders extends ToLineItemTransformation {
heights.sort((a, b) => b - a); heights.sort((a, b) => b - a);
heights.forEach((height, i) => { heights.forEach((height, i) => {
const headlineType = headlineByLevel(2 + i); const headlineLevel = i + 2;
parseResult.pages.forEach(page => { if (headlineLevel <= 6) {
page.items.forEach(item => { const headlineType = headlineByLevel(2 + i);
if (!item.type && item.height == height && !isListItem(item.text())) { parseResult.pages.forEach(page => {
detectedHeaders++; page.items.forEach(item => {
item.annotation = DETECTED_ANNOTATION; if (!item.type && item.height == height && !isListItem(item.text())) {
item.type = headlineType; detectedHeaders++;
} item.annotation = DETECTED_ANNOTATION;
item.type = headlineType;
}
});
}); });
}); }
}); });
} }