Thank you for your post.
Based on this part of the message
"write.lock, _0.cfs, _0.cfx, _0.fdt, _0.fdx, _0.tvd, _0.tvf, _0.tvx"
, it looks like both of your problems are related to content classification engine (Lucene). You are seeing this problem for both barcode and searchclassification (content based classification) because in community edition, all classification options are enabled by default regardless of document assembler plugin.The short answer to this problem is to delete files that are locking the lucene indexes. Follow these steps:
1- stop the server
2-Delete all the files in the lucene folder (C:\Ephesoft\SharedFolders\learn-index\2) except _0.cfs, _0.cfx, segments.gen, segments_2
3- restart the serverThe long answer is as follows:
Content classification uses luecene to learn classify the documents. The idea is that you give sample documents for the documents types that you wish to classify with content classification. For example, for batch class id 2, the samples are located C:\Ephesoft\SharedFolders\LuceneSearchClassificationSample\2
There are also sample folders for image comparison/classification in C:\Ephesoft\SharedFolders\ImageClassificationSample\2
After giving the samples, you are supposed to go to the admin module and hit the "Learn" button. This button generates hOCR files (from tif files) for content classification and creates thumbnails for the image classification. hOCR files are than used to build lucene indexes. During this process lucene LOCK files may be created but these files should be cleaned automatically by Ephesoft. For some reason, in your case, these files are still locked so lecene is not able to classify the documents you are importing.
You can see these files in the C:\Ephesoft\SharedFolders\learn-index\2 folder.
We will create a bug report for this but if you can replicate the issue and tell us the steps, it would be a great help.
Please note that, in the next version, we will change the folder structure so that every batch will have a main folder for all the settings such as search classification or image classification.
Also, please note that all the folder paths in the system are configurable. Above folder names are for default community installation.
Also, searchclassification, contentclassification, lucene classification are all same things. We give samples and we train them using lucene indexing. This allows the Ephesoft system to classify and separates the documents based on lucene ranking of each page. This is how we eliminate separator sheets and save labor and trees.