i need to index pdf files in my laptop running windows 7

Today (Saturday), I decided to try an experiment based on what I found Thursday evening and posted above.

This was my setup:

1.    WinXP desktop & Win7 x64 laptop both cabled to router ports (no wireless).

2.    (All remaining steps were performed from the Win7 x64 laptop) Started the Indexer Status Desktop Gadget and set it to fast-forward (higher system priority to indexing, although I don’t know yet if this step is absolutely necessary).

3.    I did not change any other Indexing Options.

4.    There are 72 sub-folders under the Documents folder. Using Windows Explorer, I started with the contents of the first sub-folder and copied no more than 10 PDF files at a time (drag-and-drop) from the WinXP desktop, across the LAN, to the corresponding sub-folder on the Win7 x64 laptop. I did not delete any of the files in any of the sub-folders on the Win7 x64 laptop before doing this, so each copy action resulted in the “Copy File” Windows popup. I clicked on “Copy and Replace” and also checked the box to perform the same action for the next 9 conflicts. This resulted in every file being re-written on the Win7 x64 laptop.

5.    These sub-folders only contained PDF type files.

6.    At no time did I rebuild the Index.

Before the copying, a Windows Search for unique character strings like account numbers, phone numbers or website names, produced incomplete, partial, almost random results.  I performed the above steps for 10 sub-folders. After all the files in each sub-folder were completely re-written, I again performed a Windows Search for the same unique character string as before. This time, the Windows Search produced exact, correct and complete search results. In other words - SUCCESS!

I can only speculate why copying a few files at a time works, but copying thousands of them spread across hundreds of sub-folders ends up with most not being properly indexed, or why re-building the index does not fix it either. Maybe it has to do with the file attributes being changed when the file is re-written. I don’t know.

I desperately need help in figuring this out and coming up with a better resolution. Copying 10 files at a time is NOT a practical solution when I have thousands of them to copy. File types other than PDF do not have this problem, at least not that I have identified.

OK, so I have uninstalled the Adobe iFilter and installed the Foxit PDF iFilter.  I get a few results, but I know they are not complete.

Sorry, but this post is going to get long. Sometimes, when asking for help in a forum like this, it’s difficult to know what background information is relevant and should be included in the post. Last evening, while continuing to work on my problem, I stumbled across something that initially I didn’t think was relevant, but now I think is closer to the cause of the issue.

In my opening post, I stated that, “These files are all financial statements downloaded from a company website.” That’s not exactly true.  This Win7 x64 laptop is new – I got it the last week of July, about 2 months ago. So, where I got these files is actually from my WinXP desktop system. I would say 98% of them (.doc, .xls, .txt, .pdf, .mp3, .jpg, etc.) were copied either over my LAN or via a USB external HDD using Windows Explorer drag-and-drop to copy them.

THIS, I now believe to be the source of the problem – mass copying. Re-building the index does not fix the problem.

I got my clue from a post in this forum topic:

http://forums.foxitsoftware.com/showthread.php?t=15132

Which in turn, referenced this forum topic:

http://social.technet.microsoft.com/Forums/en-US/w7itproui/thread/0faf282a-0c14-48d3-be25-bd4440b5e829

User Markd2100 posted:

Thanks Nano Warp. If I do not specify the file type and since I have told it to search for a word (that may occur inside a file, that search has been instructed to index that type file and its contents) it should return the results without me specifying. That said, it does not, and it always has.

The key factor here seems to me is that is if I create a new file or "re-save" an old file, then a search for a word inside a file (whatever types I have selected in configuring search, i.e. doc; pdf or whatever) that is new or "re-saved" is found, ie search works properly.

So, either the index itself needs to be reset in some way, or all the file attributes have to be changed in a way to make search "re-look" at that file.

Using the method specified above unfortunately does not help.

I found the same result as he did in the bolded, underlined section. His description of his problem through the rest of that topic most closely matches the problem on my PC.

Also, several weeks ago, I download and installed this Win7 Indexer Status Gadget:

http://gallery.live.com/liveItemDetail.aspx?li=45c441a6-edbc-454d-9262-d264728d6a3a&bt=1&pl=1

Clicking on the fast-forward button does a Registry edit that increases the priority of the Windows indexing service, as explained here:

http://www.neowin.net/forum/topic/613825-can-i-force-search-indexing-to-high-priority/

So, this morning, I ran this test on a sub-folder under Documents that contains 46 files related to my American Express credit card (44 monthly statements, 2 letters, 46 files total, all .pdf files).  I know that the unique credit card number appears only in the 44 monthly statement files.

1.    Used Windows search to find the unique credit card number - 2 files found.

2.    Used Windows Explorer from the Win7 PC to copy all 46 files from the WinXP desktop to the Win7 laptop over my LAN, both PCs cabled to router ports (no wireless).

3.    Repeated step 1 – 42 files found this time, but not all 44.

4.    Repeated steps 2 and 3 with the same result – 42 files found, but not all 44.

5.    Started the Indexer Status Gadget and clicked on the fast-forward button to increase the indexing priority.

6.    Repeated step 2 and 3 – this time, it found the credit card number in all 44 files. SUCCESS!

This leads me to the conclusion that there is some kind of speed/timing/race condition being reached when copying large numbers of PDF files. The indexing does not seem to be able to keep up.  Again, re-building the index does not fix the problem.

It seems the only way to achieve reliable indexing of PDF files being mass copied from another source, is to do the copying in small groups of files, with the help of the Indexer Status Gadget increasing the priority of the Windows indexing service. This is a totally impractical solution when working with thousands of PDF files!

By the way, here is the configuration of my laptop: lenovo T410, i7-620M, 4GB, 320GB 7200rpm, Bluetooth, DVD-RW, Win7 Pro x64 current with all critical MS updates.

Does this new information give anybody a clue as to what might be going on here or more importantly, how to resolve it?  I don't like the prospect of having to copy over files a few at a time when I have thousands of them.