GSE search on verb lemma agreement

[Pere Casanellas]

I am trying to search all the verses in the WTM version which have two verbs with the same lemma.

Could anybody explain to me why the attached GSE search does not work? For example, it finds Gen 1:1 and Gen 1:2, where there are no two verbs with the same lemma.

Thanks in advance for your help.

Pere

[David Kummerow]

Try the attached.

Regards,
David.

[Pere Casanellas]

Dear David.

Your search works fine. Thank you very much. I see that the only significant difference between your search and the one I had sent to the forum, is that you have added the Ordering and Proximity Options Window. Please, could you (or anybody else) be so kind as to explain why or in which cases it is necessary to add this window? I thought that it was not necessary, if word order, number of intervening words and puntuation, etc., was not important.

Pere

[Philip Brown]

Hi, Pere,

I think there is a glitch in the GSE that is requiring the proximity specification.
1. The WTM search .#1@v* #1@v* generates the correct hits from the commandline
2. When I export the above search to the GSE, it creates precisely (I think) the query you created that does not work.

Apparently without the proximity specification, the GSE doesn't know that the two lemmas have to be distinct. Why the commandline does know that, I don't know.

[Pere Casanellas]

Thanks, Philip!

Your comparison with the command line search is significant. In any case, in my computer the command line searches .#1@v* #1@v* and ;#1@v* #1@v* do not work fine: I get only 520 vv (beginning with Gen 2:16,17...)while the GSE search David prepared finds 2386 vv (beginning with Gn 1:1,3,5,6...).

I agree that there is some bug in the program about these searches.

Regards,

Pere

[Philip Brown]

1. The CL apparently regards any lemma agreement searches (#1 #1) as linear/string searches. All the hits I get (696) involve the same lemmas side-by-side.

2. The following linear CL search ;#1@v* *100 #1@v* yield 4928 hits, which is the correct number. Exporting it to the GSE, creates a query with AT MOST 100 words in between the two lemmas, and that allows punctuation to intervene.

3. David's query excluded any hits that had either a soph pasuq or a maqqef intervening, yielding 2386 hits. If you allow punctuation (which you should), you get the 4928 hits.

4. Allowing a soph pasuq does not result in the query crossing verse boundaries. This happens only if you have the Cross vs. button at the bottom of the query turned ON. If you do turn it on, the query finds 14110 verses!

[David Kummerow]

4. Allowing a soph pasuq does not result in the query crossing verse boundaries. This happens only if you have the Cross vs. button at the bottom of the query turned ON. If you do turn it on, the query finds 14110 verses!

Thanks, that's good to know. I hadn't run various options of the query enough to determine that.

Regards,
David.