View Issue Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0004742 | JEDI VCL | 00 JVCL Components | public | 2009-04-12 19:39 | 2009-04-15 21:23 |
Reporter | sachse | Assigned To | AHUser | ||
Priority | normal | Severity | major | Reproducibility | always |
Status | resolved | Resolution | fixed | ||
Platform | independent | OS | independent | OS Version | independent |
Product Version | 3.36 | ||||
Target Version | Fixed in Version | 3.37 | |||
Summary | 0004742: "JvCsvData - Internal Limit of MAXCOLUMNS (120) reached. CSV Data has too many columns." | ||||
Description | The testfile has 78 columns and 1300 lines (records). The problem occurs because JvCsvData expect <CR> as record delimiter. But the file contains only <LF> as delimiter usually in Unix/Linux. | ||||
Steps To Reproduce | Use a file which is cols x rows > 120 an contains only <LF> | ||||
Tags | No tags attached. | ||||
|
Following fix works for me: JvCsvData.pas:1209 ----------------- snip --------------- { if CrFlag and LfFlag and not QuoteFlag then Break; } //CR+LF OR LF if (((CrFlag AND LfFlag)OR LfFlag) and (not QuoteFlag)) then Break; ----------------- snip --------------- |
|
With boolean arithmetic you get: if (((CrFlag AND LfFlag)OR LfFlag)... => if LfFlag and not QuoteFlag |
|
Fixed in SVN. The ReadLine method now accepts all three: CR, LF, CRLF. And it now does not include the line break char in the result string where it was cut but the called anyway. |
Date Modified | Username | Field | Change |
---|---|---|---|
2009-04-12 19:39 | sachse | New Issue | |
2009-04-12 19:42 | sachse | Note Added: 0015406 | |
2009-04-13 09:39 | AHUser | Note Added: 0015408 | |
2009-04-15 21:23 | AHUser | Note Added: 0015421 | |
2009-04-15 21:23 | AHUser | Status | new => resolved |
2009-04-15 21:23 | AHUser | Fixed in Version | => Daily / SVN |
2009-04-15 21:23 | AHUser | Resolution | open => fixed |
2009-04-15 21:23 | AHUser | Assigned To | => AHUser |