Hi Vicent, Thanks for the explanation for the FOP behavior that I am noticing. In the testcase that I sent out earlier, there are some situations where the word starts right on the border (for example, in 'on an employee's' text, the last word start right on the border). Regardless of the total-fit algorithm, I would have expected the word to start on the next line instead of beginning on the table border.
Our data contains all kinds of languages and special characters and it is difficult to control what comes to us. We definitely need to go with the approach where the word flows to the next line if it cannot fit within the table-cell borders. The current total-fit algorithm does not bode well with our system and I would like to make changes in my local copy to get the desired behavior (as rendered by old FOP). If you can point me in the right direction to accomplish this, I would greatly appreciate. Thanks, Kumar Puppala -----Original Message----- From: Vincent Hennebert [mailto:[EMAIL PROTECTED] Sent: Friday, November 02, 2007 6:13 AM To: [email protected] Subject: Re: hyphenation problem Hi Kumar, Sorry for the delay, looks like I ended up forgetting this thread. Puppala, Kumar (LNG-CON) wrote: > Hi Vicent, > I am not sure if I understand you. If you look at newFop.pdf(from > previous mail), the 14th line in the table cell ( 'sion scheme') has the > overflow. It looks like the word 'scheme' started at the very edge of > the table cell and the hyphenation did not occur. The oldFop.pdf file > (generated using 0.20.5) demonstrates the correct behavior. The word 'scheme' cannot be hyphenated, so it can only be placed on one single line. But as it is wider than the column width, it would overflow the column anyway. What you see and which looks strange is a result of the total-fit algorithm: FOP optimizes line breaks for the whole paragraph, instead of working one line at a time. If it chose to put 'sion scheme' on one line, that's because that leads to an less bad overall result than if it had cut it. That's the best it can do in such a special situation. Really, you should make sure that the column is at least wide enough to make all the syllables fit on one line. > The testcase does come from real-life usecases. In some of our usecases, > we do have table cells that accommodate only 1 character (which is > usually a number). Ok, but if you actually put only one character in such cells, you won't have any problem. HTH, Vincent --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
