Using Newspro, there was an option to suppress headers for search groups when using Xpat.
I use multiple servers, but when using Xpat, I only want to use 2 or 3 servers, because the other servers dont support Xpat.
I now use UE (great program by the way), but when useing Xpat, UE uses all the servers, creating some errors. I looked through all the settings, but cant find anything like the setting in Newspro. have I overlooked something, or is this simply not possible in UE, and if not, is it possible to add this feature?
Suppress headers for search groups!
in newspro there was an option "ignore suppress headers for xpat groups"
it means if you checked it it will issue XPAT command to all servers even if they have "don't get headers" checked.
so you kept it unchecked in newspro and didn't get headers from some servers.
in UE i thought the strategy is to use maximal number of servers to get full xpat results. you also can choose to ignore errors from other servers, it shouldn't influence performance. also in UE you can attach only servers which have xpat enabled (but then differently from newspro other servers won't be used for this group so it will download bodies only from these servers as well).
but why you don't use search service instead of xpat? it is much faster and it is even safe and private (communication is encrypted, and communcation is truly encrypted for registered version, truly means even given the source code of UE and of the indexing server - without possessing your serial the "adversary" cannot decrypt it and the serial even cannot be found on the server computer).
as to the option i don't know, by now almost all newspro options made into UE and users of other newsreaders push even for more options; also as i understand what you really want is a separate option on per server basis not to get headers only for xpat groups, the option wasn't in NewsPro in NewsPro the intention was the opposite which in UE is the default.
if you tweaked "don't get headers" option in newspro only while performing xpat in UE and you want to avoid get header errors - use msg-id direct newsgroup type, detach servers, then make xpat search and then attach servers again, or you may just choose to ignore errors from non supporting servers (again, there is no performance impact and the server will give an error like 502 command not recognized or something like that right away).
in short i think to ignore the server errors (not to pay attention to them) is the best option if you want to download bodies from non-supporting servers, if you want to use only xpat supporting servers for bodies just detach other servers from the newsgroup.
it means if you checked it it will issue XPAT command to all servers even if they have "don't get headers" checked.
so you kept it unchecked in newspro and didn't get headers from some servers.
in UE i thought the strategy is to use maximal number of servers to get full xpat results. you also can choose to ignore errors from other servers, it shouldn't influence performance. also in UE you can attach only servers which have xpat enabled (but then differently from newspro other servers won't be used for this group so it will download bodies only from these servers as well).
but why you don't use search service instead of xpat? it is much faster and it is even safe and private (communication is encrypted, and communcation is truly encrypted for registered version, truly means even given the source code of UE and of the indexing server - without possessing your serial the "adversary" cannot decrypt it and the serial even cannot be found on the server computer).
as to the option i don't know, by now almost all newspro options made into UE and users of other newsreaders push even for more options; also as i understand what you really want is a separate option on per server basis not to get headers only for xpat groups, the option wasn't in NewsPro in NewsPro the intention was the opposite which in UE is the default.
if you tweaked "don't get headers" option in newspro only while performing xpat in UE and you want to avoid get header errors - use msg-id direct newsgroup type, detach servers, then make xpat search and then attach servers again, or you may just choose to ignore errors from non supporting servers (again, there is no performance impact and the server will give an error like 502 command not recognized or something like that right away).
in short i think to ignore the server errors (not to pay attention to them) is the best option if you want to download bodies from non-supporting servers, if you want to use only xpat supporting servers for bodies just detach other servers from the newsgroup.
Thank you Alex, for your quixk reply.
I use the search service most of the time, but sometimes, and I mean sometimes, files are missing or incomplete using search (I used to use Grabit to do this, but switched to UE), and for those times, I used Xpat which gives me better results.
If there is indeed no performance impact, and the errors are the only things the servers give, than I guess I will go for the last option, just ignore the errors and enjoy the program.
Thanks again for your support.
I use the search service most of the time, but sometimes, and I mean sometimes, files are missing or incomplete using search (I used to use Grabit to do this, but switched to UE), and for those times, I used Xpat which gives me better results.
If there is indeed no performance impact, and the errors are the only things the servers give, than I guess I will go for the last option, just ignore the errors and enjoy the program.
Thanks again for your support.
the search service is fed from 3 news servers, so it is unlikely you'll get missing results unless they are really missing. also the server has been designed to survive such events as e.g. system going down due to power outage, in short the database state on the disk at any moment is consistent (it is also the case in UE but for the server it is even more strictly implemented).
the main thing you should narrow the search right so the number of results will be less than 10000 AND dates/age will cover the span of the post.
if you get 10000 results - you get random results (not ordered from date to date) to provide more coverage (you can always define date range to get less), then you narrow it e.g. by defining more exact subject or author, but you should be careful with the span since if post was over long time you may miss the beginning or the end of the post (also date might be sporadic sometimes some parts posted later).
if you see any missing results let me know i'll check it here, it shouldn't be the case, because it is unlikely all 3 servers will change header range or will add the same articles out of order in the same group.
the main thing you should narrow the search right so the number of results will be less than 10000 AND dates/age will cover the span of the post.
if you get 10000 results - you get random results (not ordered from date to date) to provide more coverage (you can always define date range to get less), then you narrow it e.g. by defining more exact subject or author, but you should be careful with the span since if post was over long time you may miss the beginning or the end of the post (also date might be sporadic sometimes some parts posted later).
if you see any missing results let me know i'll check it here, it shouldn't be the case, because it is unlikely all 3 servers will change header range or will add the same articles out of order in the same group.
I haven't missed anything yet since using UE. It was the case when I used the search engine in Grabit before using UE.
When I missed something, or things where incomplete, only then I used xpat in Newspro to get the best results.
So for now no problems, but if they come up, you are the first one to get noticed
Thanks again for your quick reply and a great program.
When I missed something, or things where incomplete, only then I used xpat in Newspro to get the best results.
So for now no problems, but if they come up, you are the first one to get noticed
Thanks again for your quick reply and a great program.