<?xml version="1.0" encoding="utf-8"?>
<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Recent changes to feature-requests</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/" rel="alternate"/><link href="https://sourceforge.net/p/hitkeeper/feature-requests/feed.atom" rel="self"/><id>https://sourceforge.net/p/hitkeeper/feature-requests/</id><updated>2007-03-15T09:35:55Z</updated><subtitle>Recent changes to feature-requests</subtitle><entry><title>Documentation on the Web site</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/12/" rel="alternate"/><published>2007-03-15T09:35:55Z</published><updated>2007-03-15T09:35:55Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net59ec7d7fbabde8502b55b6914447801049043dfd</id><summary type="html">It would be nice to put the HK documentation on-line, i.e. it should be availble as PDF and/or HTML without the need to download the whole package.

For instance, the sourceforge tab "Documentation" can be used for that purpose.

Thanks,
Heinz</summary></entry><entry><title>Intercept error if MyHits website is down</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/11/" rel="alternate"/><published>2006-08-03T11:09:50Z</published><updated>2006-08-03T11:09:50Z</updated><author><name>Joerg Hau</name><uri>https://sourceforge.net/u/jhau/</uri></author><id>https://sourceforge.net1a6cd37df9d3ea409da053223fd82c4c33e5b2af</id><summary type="html">Marco,

when the MyHits website is down for maintenance,
HKReader \(= called w/o options\) dumps the whole webpage
to stdout. Looks a bit confusing ;-\)

The relevant error message is still present in the last
line \("ERROR   Cannot obtain info from server\!\!\!\!"\),
but I think the display could be cleaned up ;-\)

à+

\- Joerg


</summary></entry><entry><title>duplicated entries in test suite!!!</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/10/" rel="alternate"/><published>2006-07-26T13:13:40Z</published><updated>2006-07-26T13:13:40Z</updated><author><name>Marco</name><uri>https://sourceforge.net/u/mpagni/</uri></author><id>https://sourceforge.net1509c105b8ef64cb518728c125255a289062e868</id><summary type="html">There is two occurences of '1C03\_HUMAN' in 
.../hitkeeper/testsuite/db/seq-test.dat
which is not permitted\!\!\!
</summary></entry><entry><title>Check reference set for run_db_test.sh</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/9/" rel="alternate"/><published>2006-07-26T12:39:29Z</published><updated>2006-07-26T12:39:29Z</updated><author><name>Marco</name><uri>https://sourceforge.net/u/mpagni/</uri></author><id>https://sourceforge.net73f2a0dc6f14fd66df217e49e06bca9188065c25</id><summary type="html">run\_db\_test.sh: 

Round One ... upload "base" data, calculate reference set

... if this produce an empty hit list \(because 
force\_compute\_current id messed up, or smethinf like
this \). it should generate a FATAL error\!\!\!\!\!\!

Marco</summary></entry><entry><title>one more switch for the tests</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/8/" rel="alternate"/><published>2006-07-21T12:10:42Z</published><updated>2006-07-21T12:10:42Z</updated><author><name>Marco</name><uri>https://sourceforge.net/u/mpagni/</uri></author><id>https://sourceforge.net5389691ab1dcbf73f2031758b01a03b89fd27969</id><summary type="html">It would be really cool to have an option to force any
database to be fully cleaned-up before launching
run\_\[db|query\]\_test\[s\]. Without this, the ouptut of
\[seq|mot\]\_\[fetch|dump\]\_nr is not reproducible in any
circumstance\!

Marco</summary></entry><entry><title>-t option propagation</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/7/" rel="alternate"/><published>2006-06-30T14:22:54Z</published><updated>2006-06-30T14:22:54Z</updated><author><name>Marco</name><uri>https://sourceforge.net/u/mpagni/</uri></author><id>https://sourceforge.netc082f5e05271dd668d5ff14dc1096f8ce7de363c</id><summary type="html">Dear Joerg,

For the launcher script otions \(e.g. hk01.sh\), could
you replace '-t' with '-x' \(for 'x'-term\). Then you
could introduce the -t &amp;lt;int&amp;gt; option to modify the sleep
time passed to HKLoader/Updater/Publisher using their
already  existing -t &amp;lt;int&amp;gt; options. This should permit
MUCH faster testing with run\_db\_tests.sh in the
'noquick' mode. 

Marco</summary></entry><entry><title>Add check in run_query_test.sh</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/6/" rel="alternate"/><published>2006-04-07T08:52:03Z</published><updated>2006-04-07T08:52:03Z</updated><author><name>Marco</name><uri>https://sourceforge.net/u/mpagni/</uri></author><id>https://sourceforge.neta20bc18f8119e24ea937d8485de67637a4007403</id><summary type="html">run\_query\_test.sh must check for empty line before
attempting to run a query and skip any line that
contains only white spaces and TABS\!

This is to faciliate the maintenance of query\_test.cmd

Marco

</summary></entry><entry><title>Parser should show "relative status"</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/5/" rel="alternate"/><published>2006-03-31T12:50:05Z</published><updated>2006-03-31T12:50:05Z</updated><author><name>Joerg Hau</name><uri>https://sourceforge.net/u/jhau/</uri></author><id>https://sourceforge.nete4fc35e1a8d56a90018b091b4273f51885c8ce45</id><summary type="html">Marco,

Here's another "nice to have": the \(Seq\)parser should
indicate the percentage of progress. 

While reading huge flatfiles, it is nice to get the
"162000 entries stored"-message, yet it would be even
more useful to see the overall \(relative\) progress. 

Suggestion:
\- print a line "entries=211104" just before the parsing
starts,
\- then something like "162000 entries \(77%\) stored",
which is updated as usual.

The only thing where I'm hesitating is that this
requires "pre-parsing" the dataset once before actually
reading the data, which may take quite a while ...

cu,

\- Joerg
</summary></entry><entry><title>features for commandline</title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/4/" rel="alternate"/><published>2006-03-31T08:21:15Z</published><updated>2006-03-31T08:21:15Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.nete6919e9d45f8925bc783eda898b67e30ad26e6b0</id><summary type="html">Hi guys,

will be possible to have the following features useful
for RKReader used in command line mode:

\- silent mode: only the last query is send to stdout
\(for pipes\!\);

\- save a query as an alias \(and not $1,$2...\), so in
command line you don't need to know the number of the
query for which you want to print the results.

Great job\!

thanks

luli</summary></entry><entry><title>merge hitkeeper and hitkeeper-data </title><link href="https://sourceforge.net/p/hitkeeper/feature-requests/3/" rel="alternate"/><published>2006-03-30T20:00:54Z</published><updated>2006-03-30T20:00:54Z</updated><author><name>Anonymous</name><uri>https://sourceforge.net/u/userid-None/</uri></author><id>https://sourceforge.net8c15d059937a586d6658dcc6b841718b27c98b3c</id><summary type="html">I just modified  the run\_query\_test script and the db/\* data set. This quite 
naturally requests for the update of the two packages: scripts+data. Why 
not keep them in sync by merging them for te release\(s\)?   Anyway, we 
have svn to track everyday changes.

Marco</summary></entry></feed>