Fix TEI-XML parsing so lines with multiple <lb> tags are handled correctly

[laurejt]

Acceptance Testing Notes

Download this sample TEI file that contains tricky edge cases for testing: sample_tei_with_footnotes.xml
Run Corpus Builder in the app to build a sentence corpus for this sample TEI file.

• Verify all sentences have correct line numbers except:
  - the one sentence on page 20 whose line begin tag has no line number attribute
  - the one sentence on page 22 which has no line begin tag at all.

---

Current TEI-XML parsing generates incorrect text when there are multiple <lb> tags on the same line.

For example, the XML line

<lb n="28"/>Das bisher Entwickelte gilt unter der Voraussetzung, daß im Fortgang der<lb n="29"/>Accumulation <hi rendition="i">das Verhältniß zwischen der Masse der Produktionsmittel und

gets incorrectly parsed as:

Das bisher Entwickelte gilt unter der Voraussetzung, daß im Fortgang derAccumulation das Verhältniß zwischen der Masse der Produktionsmittel und

[rlskoeser]

Re-opening because I'd like the team to test. @tanhaow does it make sense to use the unit test fixture for testing this?

[tanhaow]

Re-opening because I'd like the team to test. @tanhaow does it make sense to use the unit test fixture for testing this?

Yes I think it's good to use the unit test fixture here, which has included all the edge cases we've found so far.
I'll add acceptance testing note to this issue soon.