dkey Posted September 4, 2017 Posted September 4, 2017 Hello Is there a simple way to import various files (more than 2000) where tabs separate 9 different fields so that each import would populate the fields as they are? The files are in UTF8 and PLAIN TEXT Importing them with the FileMaker function : FILE->IMPORT->TEXT FOLDER simply brings each file in one records wich is useless to me as I then need to script them to separate each chunk into a different field Importing each text file manually is impossible given the number of files in the folder Each file is properly formatted therefore there would be no "empty" fileds in the import process Any help would be highly appreciated Dan
comment Posted September 4, 2017 Posted September 4, 2017 (edited) 41 minutes ago, dkey said: Importing them with the FileMaker function : FILE->IMPORT->TEXT FOLDER simply brings each file in one records wich is useless to me as I then need to script them to separate each chunk into a different field It's not entirely useless, because you can either parse them individually (as you say) or tell a script to export each file in turn and import it into the target table. Another option is to use AppleScript to automate the import of the original files. Edited September 4, 2017 by comment
dkey Posted September 9, 2017 Author Posted September 9, 2017 On 9/4/2017 at 9:36 PM, comment said: Another option is to use AppleScript to automate the import of the original files. Thanks for your reply ... unfortunately I don't know how to write the script and the files to import are about 2000 and import themmanually is very though although it could be done .... The db has 8 text fields and one Timestamp field (Media Date) tyhe order is exactly the same for each field in the TAB separated files as follows: SCNSEQTAKE Scene Seq Take Duration Name Media Date Media Folder FileHash Thanks a lot for your help
comment Posted September 9, 2017 Posted September 9, 2017 Why don't you import them as files in folder and parse out the data to individual fields? That would be really easy (I am assuming each file holds one record only). All it would take is 9 Set Field[] steps in a loop over 2000 records, roughly: Import Records Go to Record [ First ] Loop Set Variable [$data; Value: Substitute ( YourTable::Text Content ; Char(9) ; ¶ ) ] Set Field [ YourTable::SCNSEQTAKE ; GetValue ( $data ; 1 ) Set Field [ YourTable::Scene ; GetValue ( $data ; 2 ) Set Field [ YourTable::Seq ; GetValue ( $data ; 3 ) Set Field [ YourTable::Take ; GetValue ( $data ; 4 ) Set Field [ YourTable::Duration ; GetValue ( $data ; 5 ) Set Field [ YourTable::Name ; GetValue ( $data ; 6 ) Set Field [ YourTable::Media Date ; GetValue ( $data ; 7 ) Set Field [ YourTable::Media Folder ; GetValue ( $data ; 8 ) Set Field [ YourTable::FileHash ; GetValue ( $data ; 9 ) Go to Record [Next; Exit after last ] End Loop Commit Records where Text Content is the target field for importing the file's content.
dkey Posted September 10, 2017 Author Posted September 10, 2017 (edited) 23 hours ago, comment said: Why don't you import them as files in folder and parse out the data to individual fields? That would be really easy (I am assuming each file holds one record only). All it would take is 9 Set Field[] steps in a loop over 2000 records, roughly: Import Records Go to Record [ First ] Loop Set Variable [$data; Value: Substitute ( YourTable::Text Content ; Char(9) ; ¶ ) ] Set Field [ YourTable::SCNSEQTAKE ; GetValue ( $data ; 1 ) Set Field [ YourTable::Scene ; GetValue ( $data ; 2 ) Set Field [ YourTable::Seq ; GetValue ( $data ; 3 ) Set Field [ YourTable::Take ; GetValue ( $data ; 4 ) Set Field [ YourTable::Duration ; GetValue ( $data ; 5 ) Set Field [ YourTable::Name ; GetValue ( $data ; 6 ) Set Field [ YourTable::Media Date ; GetValue ( $data ; 7 ) Set Field [ YourTable::Media Folder ; GetValue ( $data ; 8 ) Set Field [ YourTable::FileHash ; GetValue ( $data ; 9 ) Go to Record [Next; Exit after last ] End Loop Commit Records where Text Content is the target field for importing the file's content. Thanks a lot for your time and kindness The script works perfectly, however each text file has more than one paragraph how to "loop" each file so every paragraph becomes a separate record? Here an example of a file although some files have more or less paragraphs This is file 0033.txt 330101 33 01 01 00:00:50:16 DSC9267.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips FA43E370D45F362A 330102 33 01 02 00:00:44:16 DSC9268.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips BF4577E2E0F8B9E6 330103 33 01 03 00:00:53:13 DSC9269.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips E0684C862923D8A1 330104 33 01 04 00:00:53:22 DSC9270.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 7499E6FC14FE6334 330105 33 01 05 00:00:56:04 DSC9271.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 89424E00BDA46984 330201 33 02 01 00:00:29:16 DSC9272.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips F625681411136E05 330202 33 02 02 00:00:35:19 DSC9273.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips E16113C4DCB7BF73 330301 33 03 01 00:00:28:23 DSC9274.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 7255CE2131EE9FE4 330401 33 04 01 00:01:00:09 DSC9275.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 5A23DC0EDEE3A5AB 330501 33 05 01 00:01:11:01 DSC9276.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips BE2403BAA6DDDAE4 330601 33 06 01 00:00:51:03 DSC9277.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 2670DFB1ED76BD78 330602 33 06 02 00:00:48:18 DSC9278.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 5EFB822868AEE245 330701 33 07 01 00:01:30:00 DSC9279.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips 309304CE359B5E9C 330801 33 08 01 00:01:37:20 DSC9280.MOV 15/7/2017 15:31:46 /Volumes/SHOWDOW_2017/OriginalClips F12D6E0680A36767 , Thanks again Edited September 10, 2017 by dkey
comment Posted September 10, 2017 Posted September 10, 2017 Is this a one-time conversion or do you need to do this periodically?
dkey Posted September 10, 2017 Author Posted September 10, 2017 Thanks again I will need to do so periodically
comment Posted September 10, 2017 Posted September 10, 2017 You need a different strategy. Let me explain how to do this manually first, then we can talk about scripting some or all parts. You will need two tables, let's call them TEMP and TARGET. The TEMP table needs these fields: Text Content (Text) sList (Summary, List of Text Content) The TARGET table will have the 9 fields used by your data files. Start by importing the files (as folder) into the TEMP table, mapping Text Content to Text Content. Next, control-click into the sList field, select Export Field Contents… from the shortcut menu and export to a known location. Import the exported file (as tab-separated file) into the TARGET table. In the end you should have about 2,000 records in the TEMP table, and many more in the TARGET table.
dkey Posted September 10, 2017 Author Posted September 10, 2017 1 hour ago, comment said: then we can talk about scripting some or all parts. Thanks again Comment ... I did everything as suggested How do I need to fix the previous script if is still the one I need to use? Dan
comment Posted September 10, 2017 Posted September 10, 2017 5 minutes ago, dkey said: How do I need to fix the previous script if is still the one I need to use? There is nothing in the previous script that's applicable to this method.
dkey Posted September 10, 2017 Author Posted September 10, 2017 (edited) I do get the 1998 files however the tabs get lost This seems to be an a regular problem with Filemaker Opening the file in my Textwrangler I just see spaces where I should find <tab> except in the last part of the paragraph where TextWrangler shows two "red" inverted question marks as such 000101 00 01 01 00:00:24:24 DSC_0661.MOV 7/8/2017 15:54:58 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 1FA556300502009F000102 00 01 02 00:00:57:12 DSC_0662.MOV 7/8/2017 15:55:52 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 200AC7FEE96A2ED4000103 00 01 03 00:00:09:15 DSC_0663.MOV 7/8/2017 15:59:50 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 F32E306D85C20ECD000104 00 01 04 00:00:50:19 DSC_0664.MOV 7/8/2017 18:01:04 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 21F2E7348A2D4B3E000105 00 01 05 00:00:52:23 DSC_0665.MOV 7/8/2017 16:02:42 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 4DC482E8DED4C943000201 00 02 01 00:00:01:02 DSC_0666.MOV 7/8/2017 16:06:18 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 CDAC4BC790AB5592000202 00 02 02 00:00:27:12 DSC_0667.MOV 7/8/2017 16:06:22 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 B1731833D7BF69B8000203 00 02 03 00:00:18:15 DSC_0668.MOV 7/8/2017 16:07:40 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 F95ABC734A63BE81000204 00 02 04 00:00:18:00 DSC_0669.MOV 7/8/2017 16:08:12 /Volumes/SHOWDOW_2017/MEILI Walks 01/010 16D9DD0074C87250 This is the first line as I see it in Text wrangler as it contains the 9 paragraphs of the first record Is there a way to originally import the files including the various tabs in The Text Content field? The original files in Text Wrangler are standard text in basic UTF8 with the MACINTOSH carrige returns. Sorry to give you all these problems Edited September 10, 2017 by dkey
comment Posted September 10, 2017 Posted September 10, 2017 Could you zip one of the original files and attach it here?
dkey Posted September 10, 2017 Author Posted September 10, 2017 Hi Comment I found a "very silly" way to fix the problem (although I am always in troubles importing text files in text fields in FMP) In TextWrangler I replaced the tabs with "ç" as it doesn't exist in any of the files, and added a "∂" before the Carriage return Then I exported to FMP using UTF8 with BOM I followed your instructions and exported the files as tab Back to TextWrangler I converted the "ç" to tab and removed the "∂" before the carriage return Importing in the Target this time placed every part in the various fields as needed Well after all I am no spring chicken but it worked so thanks again for your very good assistance Kind regards Dan PS I still wonder how to find a proper way to import text into FMP keeping things as they originally are --- I've had all kind of problems with it and never found a proper solution If you like I will add an original file for you ... hoping I dont abuse your time Here it is 00.txt
comment Posted September 10, 2017 Posted September 10, 2017 (edited) 11 minutes ago, dkey said: I followed your instructions and exported the files as tab That's not following my instructions. I suggested exporting the summary field's contents using the Export Field Contents… command. If you exported the records to a tab-delimited file, then of course the in-field tabs were converted to spaces: http://www.filemaker.com/help/16/fmp/en/#page/FMP_Help%2Ftab-separated-text-format.html%23 Edited September 10, 2017 by comment
dkey Posted September 10, 2017 Author Posted September 10, 2017 I believe I did this is the visual sequence of my action Import file: result of the import in TEMP Table Export sList Table Text Wrangle page after the exportt Import file into TARGET Table Result in the TARGET Table after importing I am not sure what I did wrong. Sorry
comment Posted September 10, 2017 Posted September 10, 2017 3 minutes ago, dkey said: I am not sure what I did wrong. Your third step (Export sList Table) is wrong. You are exporting records as tab-separated file. You should be exporting field contents instead. Please read this carefully: 3 hours ago, comment said: Next, control-click into the sList field, select Export Field Contents… from the shortcut menu and export to a known location.
dkey Posted September 10, 2017 Author Posted September 10, 2017 Thanks I got lost previously now everything is oK Thanks and regards
Recommended Posts
This topic is 2629 days old. Please don't post here. Open a new topic instead.
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now