Configurable raw output format for query results

[Λύχνις Δαν]

Hi ya,

 

This comes up from time to time and it came up the other day again. I have been meaning to write up a proper feature request on this for a little while. So here it is. This is not no work but it would really help a number of use cases.

 

Currently if one does a search and gets an analysis window of the output words, it is pretty configurable. The same is true of the parsing output. It's also true to some extent of the Concordance output. But in all these cases the output it the relevant pane is in one way or another polished a bit. For example it is indented, or counts are preceded by an "=", or the output is in rows under subheadings of one kind or another. I'm speaking here primarily of the Analysis output. One has output like this :

 

Total number of verses = 1156
(total number of verses displayed = 1156)

θεός (1317 total words)

Number of different forms = 14:

θεός god, God = 1317
θεέ = 2
Vocative = 2
θεοὶ = 4
Nominative = 4
θεοί = 1
Nominative = 1
θεοῖς = 1
Dative = 1
Θεὸν = 1
Accusative = 1
θεόν = 65
Accusative = 65
θεὸν = 82
Accusative = 82
θεός = 64
Nominative = 59
NominativeNominative
VocativeVocative = 5
θεὸς = 245
Nominative = 239
NominativeNominative
VocativeVocative = 6
θεοῦ = 691
Genitive = 691
θεοὺς = 2
Accusative = 2
θεῷ = 159
Dative = 159


This is great information and handy if this is all you need. If however you need to import it into another tool it requires no small effort to clean up and process. Code must be written to remove blank space, insert delimiters, headings must be removed, and where necessary information must be duplicated and inserted into rows.

 

In the Parsing output you can get this :

 

Matt. 1:23
ἰδοὺ ἰδού (εἶδος) Interjection note! pay attention! look!
ἡ ὁ Article feminine singular nominative the, who, which
παρθένος παρθένος Noun feminine singular nominative virgin
ἐν ἐν Preposition (+dative) (+dat) in, with, by, to
γαστρὶ γαστήρ Noun feminine singular dative stomach, womb
ἕξει ἔχω Verb third singular future active indicative to have
καὶ καί Conjunction coordinating (copulative) and, also, even, and yet, but
τέξεται τίκτω Verb third singular future middle indicative to bear children
υἱόν υἱός Noun masculine singular accusative son
καὶ καί Conjunction coordinating (copulative) and, also, even, and yet, but
καλέσουσιν καλέω Verb third plural future active indicative to call, summon, invite
τὸ ὁ Article neuter singular accusative the, who, which
ὄνομα ὄνομα Noun neuter singular accusative name
αὐτοῦ αὐτός Pronoun (personal) third masculine singular genitive he, she, it; self, same; they (when pl)
Ἐμμανουήλ Ἐμμανουήλ Noun masculine singular accusative (proper) Emmanuel
ὅ ὅς Pronoun (relative) definite neuter singular nominative who, which, what
ἐστιν εἰμί (εἰμί_1) Verb third singular present indicative to be, exist (sum)
μεθερμηνευόμενον μεθερμηνεύω (μετά, ἑρμηνεύω) Verb present middle participle neuter singular nominative to translate, interpret
- μεθερμηνεύω (μετά, ἑρμηνεύω) Verb present passive participle neuter singular nominative to translate, interpret

 

This simple paste looks worse but it's actually not so bad. It has less formatting. But nonetheless there are heading for each verse, blank lines and so on.

 

Now the above formats have utility for humans to read but for machine processing they are less than straightforward if subsequent machine processing is in view.

 

With that preamble then, I would like to request a configurable raw output format be made available. It would ideally be able to:

 

save in a very simple delimited format like Comma Separated Values,

only produce rows of data with columns as requested,

empty columns in any given row would still be indicated by adjacent delimiters

no indentation, =, bracketing and such

quoting only sufficient to escape field separators in the column data

no headings though perhaps an option to have one heading line naming the columns

Unicode - of course

configurable delimiter

control of sort columns, though honestly this is not a big issue in most cases.

 

As an example then how would the analysis output above be redone in this format. Like this perhaps :

θεέ,Vocative,2
θεοὶ,Nominative,4
θεοί,Nominative,1
θεοῖς,Dative,1
Θεὸν,Accusative,1
θεόν,Accusative,65
θεὸν,Accusative,82
θεός,Nominative,59
θεός,NominativeNominative,
θεός,VocativeVocative,5
θεὸς,Nominative,239
θεὸς,NominativeNominative,
θεὸς,VocativeVocative,6
θεοῦ,Genitive,691
θεοὺς,Accusative,2
θεῷ,Dative,159

 

In a similar way the parsing output about would duplicate the chapter and verse information on each row. BTW, rather than supporting numerous canonical forms of book, chapter, verse, just put each in a column. It's easier to take that and assemble what you want than have to parse forms first and then do it anyway. It also helps sorting.

 

I totally get it if there is not wide support for this idea. Yet Accordance is a formidable product that turns the texts and supporting works into databases. I would love to be able to work a little more directly with those databases at times if possible. If there is any interest in this I'd be happy to discuss the idea further.

 

Thanx for your consideration.

Daniel

[Michel Gilbert]

+1

[Joe Weaks]

I don't think there is wide support. Most uses require the nicely visual display of the current window and that's all.

When I requested this years ago, I suggested adding a "Save as... tab delimited file" as an option.

[EricC]

+1