U̠t-Ma'in [gel] FLEx Wordlist database README
This readme is specially written so the two datasets mentioned below can be merged. The request is that DB1's content is merged into DB2. The older DB's content is mereged into the newer DB, but we hold that the older DB is actually more authoritative.
The two data bases
There are two databases which need to be merged with all data and relationships being retained. The are named as follows: FLEx backup filename:
- gel Wordlist 2013-02-22 1819.fwbackup (DB1)
- gel 2017 words and texts 2017-07-12 0021 Backup to send to Dallas for merger.fwbackup (DB2)
The (DB1) FLEx Database contains data imported from an .xls; 5 custom fields were added. Semantic Domain information from a built-into-FLEx list is also matched to each entry. We seek to retain all of this information in the merged dataset.
FLEx 8.0 will open (DB1). When it does it converts the file to "the most recent version". So this file is older than FLEx 8.0 but we are unsure what version of FLEx 7 was used to create the file.
The DB2 dataset is created with FLEx 8.3.8 on Linux. It too has custom fields. DB2 has additional lexemes (new entires compared to DB1) that come mostly from texts processed in DB2.
gel_ModOrth
Writing systems for (DB1)
Vernacular:
- gel_IPA
- gel_orth - no data from this writing system in gel Wordlist 2013-02-22 1819.fwbackup should overwrite gel_orth in the gel 2017 words and texts 2017-07-12 0021 Backup to send to Dallas for merger.fwbackup.
Analysis:
- Eng - English
- Swa - Swahili
- Frn - French
- Hau - Hausa
- Hau_CL - Hausa Common Language
- Spn - Spanish
Fields used in (DB1)
Custom fields:
- Plural
- Class Number
- Class
- POS fr List
- WACWL
Other fields used:
- Lexeme
- Citation Form
- Pronunciation
- Semantic Domain
- General Note
- Note
- Gloss
Comparison of fields
The following table shows the custom fields from the two databases and how alignment should be done. The table below can also be viewed in Google spread sheets here: https://docs.google.com/spreadsheets/d/1xfwVEBs5TKpG6g37ZSOzpzjQKWopbNowqVNx1m4tKEU/edit?usp=sharing
(DB1) data can overwrite lexical data in the other FLEx database (DB2), as long as the lexemes are matched to the data in (DB2) on the number from the WACWL custom field. Both Databases should have WACWL numbers in them. DB2 has leading zeros on this numeric field. We wish to keep these leading zeros in the final output.
| Custom Fields | Writing System | Custom Fields | Writing System | |
|---|---|---|---|---|
| (DB1) gel Wordlist-2013-02-22 | Notes on DB1 | (DB2)gel 2017 words and texts - Linux - Send/Receive | ||
| Class | gel_IPA | Agreement Pronoun Sg | gel_IPA | |
| Class Number | Eng | |||
| Plural | gel_IPA | |||
| POS fr List | Eng | |||
| WACWL | Eng | WACWL numbers under 1000 have no leading zero. SIL 1700 Wordlist has the leading zero. I'd like to keep the leading zero. Otherwise the numbers are the matching data point for each lexical entry. | SIL 1700 Wordlist | Text Eng |
| Tone Study 2013 | ||||
| Researcher | List Reference | |||
| Plural Citation Form | ||||
| Hausa Elicitation blame | ||||
| Grammatical Category of Analysis Language | ||||
| Agreement Pronoun Pl | ||||
| 1700 appendix class marker field (temp) | ||||
| 1700 appendix red field (temp) | ||||
| Tone Study Singular Lexeme | ||||
| Tone Study Plural Field | ||||
| Import instance source | ||||
| SJ Tone field |