White Peak Software Forums / Official Support Forums / Killink CSV Support / Support for Tab Seperated files lost in last version. / Latest Posts

RE: Support for Tab Seperated files lost in last version.

Kirby Turner — Fri, 22 Jun 2007 08:57:29 GMT

[quote][b]Kirby Turner (6/22/2007)[/b][hr]A new official release will be published later today.[/quote]

The new release is available from the product web site. Here's the link to the download page:

http://www.whitepeaksoftware.com/killink-csv-download.aspx

-KIRBY

RE: Support for Tab Seperated files lost in last version.

Kirby Turner — Fri, 22 Jun 2007 04:25:06 GMT

[quote][b]Traderhut (6/21/2007)[/b][hr]Well, one of the fields is the keywords to use for the product, and that is in the standard format of 'Keyword,keyword' - and I didn't use to support having "'s in the file. (Something I gave up on and wrote the code as Excel stuffs them in, and do does (did?) your software. I don't see why I need quotes in the file if it is 1PensFred Flintstone123 pebbles way... The fields are all separated by tabs, and as long as they don't contain any, who needs the quotes?
[/quote]

Have you tried the quoting rule "Value contains NULL, quote, or delimiter characters"? This quoting rule will only quote the field if the value contains a NULL character, a quote character, or a delimiter character.

The quoting rule is available at Tools > Options, or press the F8 key, under File Type Options > Default Settings.

I also plan to add a new quoting rule very soon that will quote when a "Value contains NULL or delimiter characters". I know this rule will definitely work for you.

-KIRBY

RE: Support for Tab Seperated files lost in last version.

Kirby Turner — Fri, 22 Jun 2007 04:18:57 GMT

[quote][b]Traderhut (6/21/2007)[/b][hr]Why is it 'auto-detecting' at all? I have the extension of '.tsv' registered (as ;tsv at the end of the list) as a Tab Separated file. It seems to me that it should just check that first, and then not try to determine the file type beyond that.[/quote]

You are absolutely right. If the file extension is registered to one and only one file type it should always use the delimiter from that file type. Killink CSV was doing this before but I accidently introduced a bug that prevented this check when I re-wrote the auto-detect logic. That bug has now been fixed.

To answer the question why have auto-detecting at all? There are a number of reasons.

1) A file extension maybe defined for more than 1 file type. Example, .txt is defined for tab and pipe delimiters by default with new installs. The auto-detect is used to determine which is the correct delimiter. If the auto-detect cannot figure it out then the user is prompted to select a delimiter.

2) The .xls file extension is for Excel files but sometimes download web sites will use this extension for comma-separated value files. In this scenarion the auto-detect inspects the file contents to determine if the file is actually an Excel file or a CSV file.

3) There are different Excel file formats. Killink CSV supports some Excel formats but not all. The auto-detect is used to determine if the Excel file format is supported.

4) The file may not have a file extension or the file extension may not be registered with Killink CSV. The auto-detect will attempt to determine if the file is indeed a delmiited text file and if so attempt to determine the delimiter type.

Overall, the goal behind using the auto-detect is to avoid having to prompt the user each time a file is opened and there is uncertainty.

Prior to version 1.3 Killink CSV would always prompt the user if the file extension is registered more than once or not registered at all. It would also blindly open any file with a .xls extension as an Excel file even if the format is not supported (causing an error) or the file contents are actually a CSV text file (also causing an error).

Version 1.3 is a major improvement in this area. And thanks to your report bug it is better today than it was yesterday.

A new official release will be published later today.

-KIRBY

RE: Support for Tab Seperated files lost in last version.

Traderhut — Thu, 21 Jun 2007 20:28:39 GMT

Thanks, I'll try it out ASAP... (It's downloading in the background)

However, I do have a question... Why is it 'auto-detecting' at all? I have the extension of '.tsv' registered (as ;tsv at the end of the list) as a Tab Separated file. It seems to me that it should just check that first, and then not try to determine the file type beyond that.

And Yes, I have a bunch of ','s on the 2nd line... Actually, the file is my product database written to tab sep. files... Why Tab? Well, one of the fields is the keywords to use for the product, and that is in the standard format of 'Keyword,keyword' - and I didn't use to support having "'s in the file. (Something I gave up on and wrote the code as Excel stuffs them in, and do does (did?) your software. I don't see why I need quotes in the file if it is 1PensFred Flintstone123 pebbles way... The fields are all separated by tabs, and as long as they don't contain any, who needs the quotes?

RE: Support for Tab Seperated files lost in last version.

Kirby Turner — Thu, 21 Jun 2007 14:57:38 GMT

Hi Chert,

I have made a fix that addresses the immediate problem you reported. I'm still testing so it will not be made available from the web site until tomorrow, but you can download the fixed version here:

~~http://www.whitepeaksoftware.com/downloads/setup-killinkcsv-116.exe~~
(This test version is no longer available.)

Thanks for reporting the problem.

-KIRBY

RE: Support for Tab Seperated files lost in last version.

Kirby Turner — Thu, 21 Jun 2007 14:01:06 GMT

The release in question (v1.3.2) uses a different approach to auto-detect a file's delimiter type and it seems you have found a file that breaks it. I'm guessing the file, or at least the beginning of the file has more comma in it than tabs which is why Killink CSV mis-detected the correct delimiter type.

I'll work on a fix right away. Look for an update sometime tomorrow (Friday).

Also if possible, will you send me a sample of the file that fails to load correctly to the email address [email]support@whitepeaksoftware.com[/email]? It doesn't have to be a complete file, just the first few lines. By the way, I understand if you cannot send the sample data as it may include information that cannot be shared.

Thanks,
-KIRBY

RE: Support for Tab Seperated files lost in last version.

Traderhut — Thu, 21 Jun 2007 13:12:57 GMT

FYI: After looking at it, it is parsing it as Comma separated, in spite of the settings.

Support for Tab Seperated files lost in last version.

Traderhut — Thu, 21 Jun 2007 13:09:47 GMT

I just downloaded the latest version (windows installer) and it opened up the last file I had (.tsv ext. - which I registered as Tab Separated- and have confirmed it is still so)

And it put all of the columns in the A cell, and most of the next line in one cell.. Pretty much useless.

I had wanted to use this version to get support for row delete, but it looks like I'll have to roll back to the old version (which was about 3 revisions back)