Re: [WebDNA] Can WebDNA corrupt a db?
This WebDNA talk-list message is from 2012
It keeps the original formatting.
numero = 108040
interpreted = N
texte = >=20 > When I've had badly formatted dbs in the past, the cause was=20 > usually a db that was installed from another source, such as=20 > a FileMaker export that was not done properly. In cases=20 > like these I have fixed or replaced the offending db file. >=20 Same for me. IMO the most important thing to be aware of is the = line-endings (Windows or Unix). If for example, you use a database that = was from a windows source moved to a unix source or vice versa, then you = can have a line-ending mismatch that will definitely cause corruption. = The thing that is tricky, from what I've seen, is that you may not be = aware of any corruption (line-ending mismatch) and the database will = seem to work fine for a long time if all it contains is fairly = straightforward strings and numbers. Then, all it may take to crash the = db is adding a record that contains a line break inputed through a = textarea. =20 If you use FTP for transferring files, be sure that your file encodings = match with the server. Some FTP clients can automatically convert files = (from up or down) to a particular encoding automatically.=20 I think because of the flat-file format, WebDNA dbs are quite vulnerable = to corruption like any other text file. Every keystroke is recorded, so = any inconsistency in character encodings or field delineation will = corrupt it. If you think a database file may be corrupt, you can replicate it and = replace it with the newly created one, all via webdna. I've used something like this... [writefile yourDbCopy.db][listfields yourBrokenDb.db][hideif = [index]=3D]
[/hideif][fieldname][/listfields] [/writefile] [search db=3DyourBrokenDb.db&neXXdatarq=3Dfind_all][founditems] [append db=3DyourDbCopy][listfields yourBrokenDb.db][hideif = [index]=3D]&[/hideif][fieldname]=3D[interpret][[fieldname]][/interpret][/l= istfields][/append] [/founditems][/search] [moveFile from=3D yourDbCopy.db&to=3DyourBrokenDb.db] [flushdatabases] This will actually work well if your corrupted db is still readable. =20 =20 I typically create and administer all dbs from webdna itself. My rule = is to never open a db in a text editor. Any time you do, you expose = your db to risk. =20=
Associated Messages, from the most recent to the oldest:
>=20 > When I've had badly formatted dbs in the past, the cause was=20 > usually a db that was installed from another source, such as=20 > a FileMaker export that was not done properly. In cases=20 > like these I have fixed or replaced the offending db file. >=20 Same for me. IMO the most important thing to be aware of is the = line-endings (Windows or Unix). If for example, you use a database that = was from a windows source moved to a unix source or vice versa, then you = can have a line-ending mismatch that will definitely cause corruption. = The thing that is tricky, from what I've seen, is that you may not be = aware of any corruption (line-ending mismatch) and the database will = seem to work fine for a long time if all it contains is fairly = straightforward strings and numbers. Then, all it may take to crash the = db is adding a record that contains a line break inputed through a = textarea. =20 If you use FTP for transferring files, be sure that your file encodings = match with the server. Some FTP clients can automatically convert files = (from up or down) to a particular encoding automatically.=20 I think because of the flat-file format, WebDNA dbs are quite vulnerable = to corruption like any other text file. Every keystroke is recorded, so = any inconsistency in character encodings or field delineation will = corrupt it. If you think a database file may be corrupt, you can replicate it and = replace it with the newly created one, all via webdna. I've used something like this... [writefile yourDbCopy.db][listfields yourBrokenDb.db][hideif = [index]=3D][/hideif][fieldname][/listfields] [/writefile] [search db=3DyourBrokenDb.db&neXXdatarq=3Dfind_all][founditems] [append db=3DyourDbCopy][listfields yourBrokenDb.db][hideif = [index]=3D]&[/hideif][fieldname]=3D[interpret][[fieldname]][/interpret][/l= istfields][/append] [/founditems][/search] [moveFile from=3D yourDbCopy.db&to=3DyourBrokenDb.db] [flushdatabases] This will actually work well if your corrupted db is still readable. =20 =20 I typically create and administer all dbs from webdna itself. My rule = is to never open a db in a text editor. Any time you do, you expose = your db to risk. =20=
aaronmichaelmusic@gmail.com
DOWNLOAD WEBDNA NOW!
Top Articles:
Talk List
The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...
Related Readings:
[WebDNA] WebDNA Roll Call (2013)
TCP Connect (1999)
[price] maddness (2002)
RE: Missing contexts on NT (1997)
WCS Newbie question (1997)
IE Cache Problems... (1999)
Search in 2 or more catalogs (1997)
Keep away (1997)
WC1.6 to WC2 date formatting (1997)
Secure server question (1997)
Multi-processor Mac info ... (1997)
Can GMT be called from the OS ? (2004)
apostrophe in search item (1997)
Special delete ... (1997)
A link and two command (1998)
File Uploads... (1997)
Location of Browser Info.txt file (1997)
Help! WebCat2 bug (1997)
Overridding E-Mailer (1998)
extending webcatalog (1997)