-
Thank you Adrian for your help. The beta 1.2 version you supplied me with resolved the sorting problems and the hint you supplied by email resolved the parsing and splitting problems. For the benefit of those that follow, here is Adrian's hint:
Continue to use Sort Wizard, but use a unique string for "Inject This Line Where Keys Change" instead of "Inject Heading Where Keys...
2009-03-12 22:16:02 UTC by stan_p_man
-
Correction to previous reply: You were obviously using the "Quit with error message" sort wizard options in the boxes for dealing with too few or too many columns, whereas I was using the "Default and continue" options. When I switched, I got the same "too few columns" error message.
The same thing happens now with the data I built with Excel. And I *know* that...
2009-03-12 14:24:52 UTC by adriancarter
-
Thanks for sending your test data. I tried it out and could not reproduce the behaviour you described. However, the data does not sort correctly according to the criteria I chose (40, 39, 41, all ascending and case insensitive). I see the same thing happening with the test data I assembled with Excel - 108 columns averaging 12-13 chars per cell, and about 8000 lines.
Do you get a complaint...
2009-03-12 13:02:29 UTC by adriancarter
-
We tried unsuccessfully to use TextWedge to split a 108 column tab delimited file consisting of about 8000 records that averaged about 1350 characters each. We attempted to use the 40th, 39th and 41st field to sort on but it did not sort on the desired fields. We were unable to sort the file using Sort Wizard (Tabbed Sort Dialog) which complained about there being too few columns. We checked...
2009-03-10 23:07:11 UTC by stan_p_man
-
The editor component recognizes UTF-8 data with BOM properly.
However, on save/split the file encoding is converted to single byte (I assume the current ANSI locale is used here).
This leads to loss of characters: Non-locale chars are convertet to "?" on save, and to spaces on split.
Will it be possible to force output encoding to a user-specified value?.
2009-03-01 15:20:49 UTC by frank_kuhnke
-
The editor component recognizes UTF-8 data with BOM properly.
However, on save/split the file encoding is converted to single byte (I assume the current ANSI locale is used here).
This leads to loss of characters: Non-locale chars are convertet to "?" on save, and to spaces on split.
Will it be possible to force output encoding to a user-specified value?.
2009-03-01 14:59:16 UTC by frank_kuhnke
-
OK, I'm starting to understand what you're trying to do. The program is behaving as I would expect. These settings are "one shot only", and when you go back into the "Configure" dialogs they are cleared.
After you set your splitting criteria in "Configure Standard Splitting", the program is set up to apply these criteria against the editor contents when you...
2009-02-12 12:23:32 UTC by adriancarter
-
Thanks for your answer. However, I did not mean two sessions between which the preferences got lost. Example:
- Open TextWedge
- Dialogs -> "Configure Standard Splitting" -> Set value "5" for "Chunks are delimited by"
- Press "OK"
- Open the Dialog "Configure Standard Splitting" again: all preferences (including "Chunks are...
2009-02-12 07:33:51 UTC by picapauamarelho
-
TextWedge only remembers these preferences for the first split you do within the current session. If you close TextWedge, these settings are lost.
Try loading the text file or pasting into the editor first, then set preferences, then do the split (or virtual split) before you close down TextWedge.
In my usage of TextWedge it would not make sense to save these preferences between sessions...
2009-02-12 07:06:14 UTC by adriancarter
-
Whatever I enter in the Dialogs "Configure Standard Splitting" or "Configure Preview" and then choosing "OK", the preferences I chose are not saved. Going back to this dialogues the default settings are shown.
Therefore, I did not succeed any split yet. Any idea?.
2009-02-10 19:08:04 UTC by picapauamarelho
