Text unit name not using ID attribute in HTML filter

[Former user]

Original [issue 65](https://code.google.com/p/okapi/issues/detail?id=65) created by @ysavourel on 2009-05-07T14:35:29.000Z:

Some HTML elements such as <p> may have an attribute 'id'. Its value should be carried to the extracted ext unit as its name (=resname in XLIFF), so some utilities can take advantage of it (alignment, update etc.)

[Former user]

Comment [1.](https://code.google.com/p/okapi/issues/detail?id=65#c1) originally posted by @ysavourel on 2009-05-08T17:41:47.000Z:

[Former user]

Comment [2.](https://code.google.com/p/okapi/issues/detail?id=65#c2) originally posted by @ysavourel on 2009-05-22T21:44:44.000Z:

Id is not valid in base, head, html, meta, param, script, style, and title elements. All other elements may have it.

[Former user]

Comment [3.](https://code.google.com/p/okapi/issues/detail?id=65#c3) originally posted by @ysavourel on 2009-05-22T22:32:23.000Z:

I am having a hard time visualizing an algorithm if we assume ID can appear on ANY element (except the above). How far way can the text be from the element and still have the ID set to TextUnit.name? What if there is not a closing element for an element that has an id? In that case we keep an id value hanging around for a long time.

The only compromise I can think of is to have a finite list of content based content elements such as p, h1, dd, dt etc.. The same elements that we consider complex TextUnits - that way the elements are \*close\* to the text they wrap and the number of possibilities is reduced considerably.

[Former user]

Comment [4.](https://code.google.com/p/okapi/issues/detail?id=65#c4) originally posted by @ysavourel on 2009-05-22T23:53:43.000Z:

The finite list sounds good enough. The useful ID would be mostly on p, hN, etc.

[Former user]

Comment [5.](https://code.google.com/p/okapi/issues/detail?id=65#c5) originally posted by @ysavourel on 2009-12-21T07:51:39.000Z:

It seems to be working now. Should we close the issue?

[Former user]

Comment [6.](https://code.google.com/p/okapi/issues/detail?id=65#c6) originally posted by @ysavourel on 2009-12-21T18:56:38.000Z: