extern/pdf-to-markdown
1
0
Fork 0
mirror of https://github.com/jzillmann/pdf-to-markdown.git synced 2024-11-27 18:23:36 +01:00
Code Issues Packages Projects Releases Wiki Activity

DetectTOC: only go for lines containing '...' words

Browse Source 
DetectTOC will work on all lines, shaving off numbers from the last of
the words in a given line, so long as the word is not all full-stops.
This implies that a TOC line is one that contains strings containing
only full-stops, and so, DetectTOC should only work on such lines.

This change will remove unwanted behaviour where DetectTOC removes
trailing numbers that we actually want to keep in lines, eg:

   Case Number : ABC 12/1234
This commit is contained in:
LoneRifle 2019-07-27 15:45:53 +08:00 committed by GitHub
parent 2869b5e5de
commit 648c0add59
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
src/javascript/models/transformations/lineitem/DetectTOC.jsx
View File

@ -30,8 +30,10 @@ export default class DetectTOC extends ToLineItemTransformation {
const pageTocLinks = [];
var lastWordsWithoutNumber;
var lastLine;
//find lines ending with a number per page
page.items.forEach(line => {
// find lines with words containing only "." ...
const tocLines = page.items.filter(line => line.words.includes(word => hasOnly(word.string, '.')))
// ... and ending with a number per page
tocLines.forEach(line => {
var words = line.words.filter(word => !hasOnly(word.string, '.'));
const digits = [];
while (words.length > 0 && isNumber(words[words.length - 1].string)) {