Bill Meyer

KDP. The challenge is that if they find any issue with a PDF after upload, the feedback is not useful. To declare "internal corruption" in a 480 page PDF which has already been verified in multiple readers is no help whatever, and leaves room only for guesswork. As of this morning, the book is again available on Amazon: https://www.amazon.com/Delphi-Legacy-Projects-Strategies-Survival/dp/B0B2TY6ZZ4 And the PDF through: https://wmeyer.tech

And that takes us back to Dalija's point. Understanding needs to precede coding.

Amazon shows the book out of print. They had a corrupted PDF, and though I uploaded a replacement almost 12 hours ago, they have not moved things forward. I am not taking the book out of print. Amazon is a thing unto itself.

UPDATE: The book is now available in PDF at: https://wmeyer.tech/books I am working on an ebook version, but the book format is not a good fit for ebook, so I am having to rework the presentation a good deal. Probably a couple of weeks till I can have it ready.

Yes, yours is fine.

I think we need to lobby the producers of plug-ins to always make shortcuts part of their options. The collision of shortcuts in plug-ins has lately become annoying, with operations I don't use hiding operations which I value highly.

UPDATE: I see that my book is available now in Europe, and some sales have been logged there. So for anyone awaiting that availability. wait no more.

Availability: Amazon informs me that it can take 3-5 days to appear in foreign markets.

I will check the settings in Amazon. This is my first publication there, and I may well have failed to check the appropriate box(es).

I will need to see whether I have made some error which prevents it being listed for your country. I expect to offer a PDF. Many books include only snippets of code. Some examples in my book are a good deal longer and would not do well in a flow format.

Not at present. I anticipate offering a PDF, but the book layout is not compatible with flowed text, so no eBook will appear.

I am building reports again with ReportBuilder after a few years away from it. At present, my report produces only one row, though there are 384 records in the dataset. I recall encountering this problem some years ago, but do not recall how I got past it. If anyone here can enlighten me, I would appreciate it.

I needed to remove the pipeline from the report, and connect it to the group header instead.

Update: I finally found the right search term, and found a small article in the rbWiki which afforded me the necessary clue. I am now getting 11 pages and all 384 records.

Thanks, I appreciate your efforts. I can't use the forums at the moment, as the license is not active, and the renewal is working its way through corporate. The creation of objects is not a big deal, the main thing is that I have a couple of columns which are always present, and then some number of groups of four columns each, and I have to allow for printer orientation affecting how many can fit, and for putting the excess onto another page (horizontal spanning.) I also handle all the column width management in code. We have only a handful of static report designs in the system.

Totally in code. We routinely need to construct dynamically, and that is the case in this one.

I have no .tmp file. Is that from a property I need to set?

Does not seem to be. Changing it produces the same result.

No filter is in use. Not sure what you are suggesting in b.... If you mean the ColumnTraversal property, it is set as ctTopToBottom, but ctLeftToRight makes no difference.

Did that, and it reports 1. However, a logging action in plMain shows 384. Still, this is at least a clue, I suppose, except I don't understand what it suggests. My log results:

The layout is relatively simple. All objects in the Detail band are created in the BeforePrint.

It is and has been.

It is, and the data in that row is from the first record of the dataset. Logging in pipeline events shows that the pipeline is active, and dataset is active and contains 384 records, and even shows that pipeline.Next fires. Once.

In a software company? That only reinforces my opinion of MBA programs.

Active members with an understanding of the relative value of polling data may simply disdain such polls, and not take the time. I can't recall a time when I participated in such a poll.