Still no Search with Hebrew Vowel Pointing??

Wyn Laidig said:

Somehow I lost the example image I had pasted in.  I will try again here:

I know it's not ideal, but you can do an analysis search, and group by result.

MJ. Smith said:

Mark, when I sort by lemma I get duplication for alternate lemmas as I expect but I also get duplications that I can't explain - giving me 4 entries for an original one. I think this is a bug but ...

Repeating rows is a standard way of displaying data that has one-to-many relationships — although I accept it's confusing.

Each of the rows are unique, so in the case of 1 Kings 20:34 there are two different lemmas each with two different roots tagged in the word מֵאֵ֨ת. Each lemma/root combination has it's own row.

This is unacceptable to me. I remember this is the very first bug that I reported inn logos 4 and at that time somebody at logos promised to look at it. But now it is already logos 5. Searching vowel points should be very easy to deal with,  just like the exact search being handled by libronix. Therefore,  it is only a priority issue and logos look like not very interested in investing time and money in perfecting the original language function.

This discourage me to upgrade my existing logos 4.Disappointing. 

Wyn Laidig said:

 I am very disappointed that Logos 5 still does not support searching with Hebrew Vowel Pointing.  

Actually, Logos5 can search on a specific word with a specific (nekudot) vowel point patterns like this:

For example say you want to Find every instance of the lexeme אֱלוֹהַּ (eloah – God) that is spelled ‘defectively’ (with just a holem, not a holem-waw) in Logos5 you could do so by:

1.Selecting the Bible search (or the Morph search) and the BHS

2. then in the search bar type in (אֱלוֹהַּ) or anything else you want with the parenthesis or without them then, search

3. I got 40 results in .11 sec

4. That should be what you are looking for but it will ignore accents.

Or, you can also do searches this way:

1.Select Bible Search

2. Select the BHS with Westminster 4.2 Morphology 

3. On the command line or search bar type in the unicode codes for the vowel points like this:

(/.*.[\u05b1\u05b2\u05b3].[\u05b7\u05b8].[\u05b7\u05b8].*/)

4. It should yield you, 859 hits

It would appear that just about everything on this page http://www.logos.com/support/lbs/HebrewRegularExpressions  will work in the Bible search, you no longer need to type in the 'marks' command or a lemma command, in fact don't do that.

You can even search for accents, too if you know their unicode values. 

We are, of course, aware of this situation. I know Design has specified a UI solution for providing even greater control over what is being matched than what LDLS3 could do (including handling accent/cantillation marks), but I don't have any information on when something like this will be implemented. Someone on the Dev side might know more.

You might be interested in knowing that the Syntax Search interface has a check box in the Text field of the Segment object 'Match marks' that behaves like the marks() operator of Libronix DLS3 - that is, with the mark checked, vowel points are respected. I know this isn't a perfect solution, since 1) it's rather fussy to set up a Syntax search just to search for a single surface form, and 2) if searching for forms with suffixes, one must be particularly careful of where the break between the segments happens (which sometimes is between a consonant and the vowel under it), since one has to specify the form for each segment - there isn't a way to specify the entire grapheme with prefixes and suffixes at one time as the surface forms are split across segments.

Has there been any update or word on the following from development?:

Vincent Setterholm said:

I know Design has specified a UI solution for providing even greater control over what is being matched than what LDLS3 could do (including handling accent/cantillation marks), but I don't have any information on when something like this will be implemented. Someone on the Dev side might know more.

BKMitchell said:

Has there been any update or word on the following from development?:

Vincent Setterholm said:

I know Design has specified a UI solution for providing even greater control over what is being matched than what LDLS3 could do (including handling accent/cantillation marks), but I don't have any information on when something like this will be implemented. Someone on the Dev side might know more.

While there is no direct capability to accomplish this, the same effect can be obtained by doing a morphological search and specifying a particular form.  Once you have specified the form, the pointing follows aside from plene or defective writing.

Thanks Mr, Setterholm!

Vincent Setterholm said:

was told this feature was up and running in a 5.2 demo (there will be a 5.1a release that goes out between now and then).

This is fantastic news![Y] As 5.1a Beta is already in use by many a user that means it won't be too long before 5.2beta is out. Finally, we'll be able to search on accent marks, and vowels in Logos. 

George Somsel said:

While there is no direct capability to accomplish this, the same effect can be obtained by doing a morphological search and specifying a particular form.  Once you have specified the form, the pointing follows aside from plene or defective writing.

Thank you Mr. Somsel![:)]

" rel="nofollow">Keep Smiling 4 Jesus said:

Logos 5.2 Beta 1 has search modifiers => http://wiki.logos.com/Search_Matching_Commands

Thanks for the heads up on this one! This is step in the right direction, indeed.

Now, I can't tell form the link if these matching commands will allow one to search on accents(and vowels) with wild cards in place of the consonants. So, for example one could search on vowel patters without specifying any particular word. And, if one could search on accents combinations without regard to the words with they fall on.  So, If I wanted to find every verse were a Zarka accent mark is preceded by a Munach accent mark could it be down with the new matching commands and if so how?