Ascii.parseInt can throw a NumberFormatException which should probably be
caught and a 400 returned.
-Tim
On Sun, Jun 5, 2011 at 4:26 PM, wrote:
> Author: markt
> Date: Sun Jun 5 20:26:37 2011
> New Revision: 1132487
>
> URL: http://svn.apache.org/viewvc?rev=1132487&view=rev
> Log:
> Parse por
**
What happens if PatternSyntaxException is thrown? (bad regex is passed in)
While it is a RuntimeException - i'd assume you'd want to log.warn(with the
regex) and return -1;
-Tim
On 6/15/2011 6:44 PM, ma...@apache.org wrote:
Modified: tomcat/trunk/java/org/apache/catalina/ssi/ExpressionParseT
The javadocs say it can be an "expensive" operation. But considering the
alternatives [and this is only done once], this seems to be reasonable.
What could be worrisome is the JDK does allow for more charsets to be added
at run-time. In which case - a user could be out of luck.
If we were scared
+1 To getting newer ideas in faster.
I guess the question to ponder is what are the future changes pending for 7?
What are the changes for the servlet spec?
I have a feeling that the servlet spec changes are more of a playground for
ideas which would bring about lots of change and potential rever
It may be overkill ... but should "absolute" be XML escaped?
-Tim
On Thu, Aug 11, 2011 at 4:13 AM, wrote:
> Modified: tomcat/trunk/java/org/apache/catalina/connector/Response.java
> URL:
> http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/catalina/connector/Response.java?rev=1156530&r1
+1
On Thu, Aug 25, 2011 at 10:44 AM, Mark Thomas wrote:
> Subject says it all.
>
> Here is my +1 to start this off.
>
>
/me confused - should this be removed from trunk instead of tc7.0.x?
-Tim
On Wed, Aug 31, 2011 at 10:31 AM, wrote:
> Author: markt
> Date: Wed Aug 31 14:31:13 2011
> New Revision: 1163643
>
> URL: http://svn.apache.org/viewvc?rev=1163643&view=rev
> Log:
> As per the comment, remove the deprecat
Wouldn't converting the whole method to use StringBuilder be more efficient
- as compared instantiating a StringWriter AND PrintWriter? We could
probably guess a good default size too of size()*40 which would assume most
headers are less than 40 chars (which is typically true except for
user-agent
I'm a little confused. On a quick glance of the method ... it appears the
method setProperty should be returning true if it did something, and false
if not. This would mean the cast is not needed (actually wrong). The fact
that the method did NOT throw an exception should mean that we should return
Looks ok ... a few comments.
RemoteCIDRFilter (most of the below apply to the valve too)
setAllow = If nothing passed - This should clear allow
setAllow = If a bad "allow" is passed - throw exception. I'd think throwing
an IllegalArgumentException is OK so no catch is needed. Depending on when
t
DefaultServlet is still using the old style debug. Its an initParameter.
So I didn't change any of that functionality.
It probably should be converted.
-Tim
Filip Hanik - Dev Lists wrote:
are we still using the debug attribute, or should it be
if log.isDebugEnabled log.debug...
[EMAIL PROTE
Unless I am misreading something - this looks to be an accident in
web.xml processing. See lines 198 - 209 in trunk for
org.apache.catalina.startup.WebRuleSet and you'll see this twice:
digester.addCallMethod(prefix + "web-app/listener/listener-class",
"addApplica
instance is being used.
-Tim
jean-frederic clere wrote:
Tim Funk wrote:
Not sure if I understand the question, the problem is our use of
digester. We had 2 listener rules being created and each called
addListener. So when an a was found - it was being
processed twice.
Yes. My idea was to
Not sure if I understand the question, the problem is our use of
digester. We had 2 listener rules being created and each called
addListener. So when an a was found - it was being processed
twice.
(Or was I missing something?)
-Tim
jean-frederic clere wrote:
[EMAIL PROTECTED] wrote:
Aut
Since there was never a release from trunk - there really isn't a change
log. When trunk is released as a new branch it should have a
RELEASE-NOTES which highlight why its a different branch but maintaining
change log for an unreleased version doesn't make much sense to me.
-Tim
Filip Hanik -
Dumb question. (More of a curiosity since I see this a lot)
Is there a reason to use:
java.io.File fd =
new java.io.File(paths[j] +
System.getProperty("file.separator") + name);
Instead of
java.io.File fd = new java.io.File(paths[j], name);
-Tim
[EMAIL PROTECTED] wrote:
It is probably due to old code which works just fine when SHA might not
have been "easily available" in all JVM's. (back in 2002?)
So a quick recap for folks ... a session id is generated by
1) Getting a random number
2) Hashing it
3) Converting the hashed bytes to something text [base64] so the
Costin Manolache wrote:
Cool. In a nutshell - I like all the ideas.
But while I like the idea of ditching Valves/LifecycleListeners - how
does this work when the component needs to work across multiple
ServletContext's? The only reason I see that Valves/LifecycleListeners
need to yet exist
Unless I am missing something - we will now have a one TldConfig per
Context. And each TldConfig will have its own tldDigester. But the
tldDigester is only used once on startup to find all the listeners.
So tldDigester only needs to live for the life of execute() where
eventType=Lifecycle.STAR
different pre-existing "bug") In which case, not much to
worry about. And since only one instance of tldDigester would exist
(instead of one per context) - its not the end of the world with respect
to memory waste.
-Tim
Mark Thomas wrote:
Tim Funk wrote:
Unless I am missing something
Mark Thomas wrote:
Tim Funk wrote:
Now I am more confused. So tldDigester is static. The init method sets
setTldValidation and setTldNamespaceAware based on the first Context it
encountered. (As well as the first StandardHost) So if another
Context/StandardHost have different settings, they are
Before reading the POST body - you should first be doing this:
request.setCharacterEncoding("UTF-8")
-Tim
André-John Mas wrote:
Hi,
I have opened issue 45957, for an issue that has bothered me for a while:
https://issues.apache.org/bugzilla/show_bug.cgi?id=45957
To resume:
Currently in Tom
If you take that form and post it - how does the server know that the
content is UTF-8? (Answer: it doesn't)
The HTML directives tell the browser to encode everything into UTF-8 on
the way to the web server. But there is nothing that tells the webserver
explicitly what the charset is of the in
If we really prefer to be particular about change logs. Then we should
create a BUGZILLA VERSION called trunk and potentially a new "product"
called tomcat 7 (or tomcat-unknown). Then any fix first goes into
Bugzilla with a description of the fix.
Then every commit message would contain the BU
Oddly enough - I started reading the code today. There are some minor
tweaks without digging too deep into the code:
ProxyConnection.java
This should be CLOSE_VAL.equals(method.getName())
if (CLOSE_VAL==method.getName()) {
PoolProperties
protected String name = "Filip Connection Pool["+(poolCou
Filip Hanik - Dev Lists wrote:
a. [ ] It doesn't belong here, take it elsewhere
b. [ ] Bring it here, we'll ship it with next major version of Tomcat
c. [X] Bring it here, we'll make it an independent module with its own
release cycle
d. [ ] I'm ok with any these option
e. [ ] Lets discuss it
1) I think you might also need in StandardHostValve.java:
setStatus(status, message);
But the more I dig the less I am convinced that the above snippet is needed.
2) And then this probably needs deprecated
public void reset(int status, String message) {
3) reset() also did this:
coyoteRe
+1 - its an evil security hole
-Tim
Mark Thomas wrote:
All,
I'd like to remove the invoker servlet entirely from trunk (and hence TC7.x
onwards) and deprecate it in 6.0.x.
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For a
On connection.close() - the pool should ensure that autocommit is reset
correctly. Example test case:
// assume autocommit for JDBC verndor is true
Connection con = datasource.getConnection();
System.out.println("con.autoconnect=" + con.getAutoCommit()\n";
// above line should say true
con.setAu
What Component should be used in bugzilla? It might be nice to add a
jdbc-pool Component. (And maybe a bayeux)
-Tim
Filip Hanik - Dev Lists wrote:
The Apache Tomcat team announces the immediate availability of
Apache Tomcat JDBC Connection Pool
Apache Tomcat Connectors version 1.0.1-beta is
This doesn't make sense to me. The Javadoc and the code check for null
to allow the use of the default value in the JVM. This patch makes all
of that invalid and sets it to true.
So either
1) this would need reverted -or-
2) Or the if (tcpNoDelay != null) needs removed - or -
3) The comments ge
Just turning the random number into a session id should sufficient and
we can forget the MD5 altogether. But if someone figures out the seed
and can guess future subsequent numbers, then they can guess future
session ids.
By using a hashing algorithm - it makes it impossible to guess what
num
IIRC - there was talk of this over a year ago. In think in the context
of that talk APR also had some duplication too and was "in scope" for
the merge.
But the dev list archives has more details and the conclusion.
-Tim
Kirk True wrote:
Hi all,
There are a number of places in NioEndpoint an
If http://host/contextpath is requested - shouldn't we be redirecting to
http://host/contextpath/ , not worrying about a null uri?
-Tim
ma...@apache.org wrote:
Author: markt
Date: Fri May 1 20:12:09 2009
New Revision: 770809
URL: http://svn.apache.org/viewvc?rev=770809&view=rev
Log:
Fix 470
Congrats!
-Tim
Mark Thomas wrote:
On behalf of the Tomcat committers I am pleased to announce that Konstantin
Kolinko has been voted in as a new Tomcat committer.
Please join me in welcoming him.
Mark
-
To unsubscribe, e-m
On a quick glance - it seems to be off of all the wikis. Including the
wiki home http://wiki.apache.org/general/
The only exception seems to be Geronimo (if we wanted a place to copy from)
-Tim
Mark Thomas wrote:
From: Konstantin Kolinko
Should there be some explicit TermsOfUse page or copyri
Remy Maucherat wrote:
The candidates binaries are available here:
http://people.apache.org/~remm/tomcat-6/v6.0.20/
According to the release process, the 6.0.20 tag is:
[ ] Broken
[ ] Alpha
[ ] Beta
[X] Stable
-
To unsubscribe
+1
I loathe while(true) (but its a personal taste) - here is a shorter non
while(true) version:
if (!aliases.isEmpty()) {
String aliasedName = name;
DirContext dirContext = aliases.get(aliasedName);
while (dirContext!=null) {
int slash = aliasedName.lastIndexOf('/');
Correct
-Tim
ma...@apache.org wrote:
Author: markt
Date: Wed Jun 10 17:28:23 2009
New Revision: 783434
URL: http://svn.apache.org/viewvc?rev=783434&view=rev
Log:
Fix typo. I assume funkman was voting for the new patch.
Modified:
tomcat/tc6.0.x/trunk/STATUS.txt
Modified: tomcat/tc6.0.x/tr
Would it be possible to just create an optional package that has all the
old class names and their implementation is just an extends of the new
version? (As well as being marked deprecated)
-Tim
On 4/30/2010 2:51 AM, Konstantin Kolinko wrote:
There is a difference in Comet API that will req
I think I am changing my mind, I am leaning towards 1.
There are cases where I think we can use the regular access log (where I
like solution 2).
But with solution 1. You can also log "oddness". Connections which
timeout, or never present a request, etc. So the format of the
CoyoteAdaptor lo
D (and possibly B)
In the case of B - instead of commenting out - wrap a filter around it
that has default behavior of not being enabled.
So it would look like this:
boolean allowTextInterface = false;
init(FilterConfig) {
String s=System.getProperty("manager.allowTextInterface");
allowTex
+1 to E
-Tim
On 5/10/2010 11:47 AM, Mark Thomas wrote:
On 10/05/2010 13:25, Konstantin Kolinko wrote:
Thoughts?
It is possible to combine A+C:
E. Use a different name for the new role that allows access to GUI
interface only,
"manager-gui", as well as new "manager-script", "manager-jmx",
"
Cool - One (very minor) nit ... A safer size might be 500.
A typical User agent string by itself is 100+ characters which will
easily take the log line over 128 characters.
StringBuilder result = new StringBuilder(128);
-Tim
On 6/2/2010 11:12 AM, ma...@apache.org wrote:
Author: markt
Date:
-0 (or +1) (Well that was clear ... wasn't it?)
The core consideration is the scope of the Tomcat project. If its to
deliver a spec compliant servlet/jsp engine. Then I would say (-0 or -1)
If the scope is to increase to also act as a repository for generic
servlets, then we are in different
Is this needed anymore in setclasspath.sh ?
# OSX hack to CLASSPATH
JIKESPATH=
if [ `uname -s` = "Darwin" ]; then
OSXHACK="/System/Library/Frameworks/JavaVM.framework/Versions/CurrentJDK/Classes"
if [ -d "$OSXHACK" ]; then
for i in "$OSXHACK"/*.jar; do
JIKESPATH="$JIKESPATH":"$i"
08/06/2010 18:14, Tim Funk wrote:
Is this needed anymore in setclasspath.sh ?
Works for me with that commented out.
Mark
# OSX hack to CLASSPATH
JIKESPATH=
if [ `uname -s` = "Darwin" ]; then
OSXHACK="/System/Library/Frameworks/JavaVM.framework/Versions/CurrentJDK/Classes&qu
> [X] Beta - go ahead and release as 7.0.0 Beta
-Tim
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
-0
Since the host name should be a valid dns name(ok it doesn't need to be
a valid dns name if is used), choosing a default which doesn't
have a valid dns name may cause woes.
But if someone came up with a fantastic alternate default name, then I
could be swayed.
-Tim
On 7/15/2010 8:51
ious configs about using host names in various places
instead of IP addresses due to relying on DNS.
-Tim
On 7/15/2010 11:11 AM, Mark Thomas wrote:
On 15/07/2010 15:16, Tim Funk wrote:
-0
Since the host name should be a valid dns name(ok it doesn't need to be
a valid dns name if is used
Not sure if this belongs on legal mailing list or not ...
Since Sun is no more and owned by Oracle ... does this page need changed
to say Oracle?
http://tomcat.apache.org/legal.html
-Tim
-
To unsubscribe, e-mail: dev-unsubs
log is protected (instead of private) is intentional?
-Tim
On 8/24/2010 6:02 AM, ma...@apache.org wrote:
+
+/**
+ * Logger.
+ */
+protected static org.apache.juli.logging.Log log
+=
org.apache.juli.logging.LogFactory.getLog(AbstractOutputBuffer.class);
---
-1 To removing spec links. We should be linking to the JCP since that is
what Tomcat implements and we WANT developers to know and download the spec.
It would be nice if we can just put the spec in as part of the docs, but
IIRC - that is against the licensing terms of the docs. If we could put
In that case ... +1 for removing links to
http://java.sun.com/products/servlets and
http://java.sun.com/products/jsp (or any http://java.sun.com page)
-Tim
On 8/26/2010 10:04 AM, Mladen Turk wrote:
On 08/26/2010 03:14 PM, Tim Funk wrote:
-1 To removing spec links. We should be linking to
+1 To intermediate page on wiki to allow newbies to understand what the
JCP is
-Tim
On 8/27/2010 6:28 AM, Konstantin Kolinko wrote:
Maybe create a Wiki page to list the relevant specification links? (I
don't have time now, but maybe later).
--
Looks good.
It would be really cool to supplement this with a YouTube video of a
screen walking through the steps laid out while playing with JConsole.
-Tim
On 9/15/2010 11:07 AM, Mark Thomas wrote:
As I mention recently, JMX is now looking pretty good. I have drafted a
blog entry [1] on th
There is no issue. If there is a typo in the developer code, there is a
typo in the code. And sometimes typos cause security issues. As a
general rule, any code which is user provided should validated and
output escaped.
This is a topic which should be discussed on the user list.
-Tim
On 9/1
Nice work ...
I think a link needs to be added to report a bug.
http://tomcat.apache.org/bugreport.html
-Tim
-
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.
FYI ...
There was a similar fix applied here but reverted ...
https://issues.apache.org/bugzilla/show_bug.cgi?id=47512#c4
http://svn.apache.org/viewvc?view=revision&revision=832792
http://svn.apache.org/viewvc?view=revision&revision=832638
-Tim
On 10/8/2010 7:57 AM, ma...@apache.org wrote:
Au
> [X] Beta - go ahead and release as 7.0.4 Beta
-Tim
On 10/15/2010 4:47 AM, Mark Thomas wrote:
The proposed Apache Tomcat 7.0.4 release is now available for voting.
It can be obtained from:
http://people.apache.org/~markt/dev/tomcat-7/v7.0.4/
The svn tag is:
http://svn.apache.org/repos/asf/
Cool. Some notes ...
(1) Since we already have
- effectiveMajorVersion
- effectiveMinorVersion
Is Context.version a "good name" to use? Since the name version is also
used by the servlet spec? Would "revision" be a less confusing name? (Or
webappRevision).
(2)
[I thought of this as a side ef
On 10/22/2010 7:39 AM, Mark Thomas wrote:
On 22/10/2010 06:27, Tim Funk wrote:
(1) Since we already have
- effectiveMajorVersion
- effectiveMinorVersion
Is Context.version a "good name" to use? Since the name version is also
used by the servlet spec? Would "revision" be a
Should we add trim()? (in case the user has a new line or spaces after
the ,)
this.resourceOnlyServlets.add(servletName.trim());
-Tim
Modified: tomcat/trunk/java/org/apache/catalina/core/StandardContext.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/catalina/core/Standar
A possible nice to have is trying to "test" whether a new config will be
OK or not.I have a patch which will allow a new arg for catalina.sh
called "configtest".
It will try to load the catalina config but log everything to standard
output. It will try to exit with a non zero status code if "b
"svn diff -x -w" version uploaded. I always wondered how to ignore
whitespace in svn ignore. (It bugged me, but never enough to rtfm)
Thanks for the tip!
-Tim
On 11/3/2010 6:46 PM, Konstantin Kolinko wrote:
Too many unneeded whitespace changes. It is hard to read. Try
svn diff -x -w
In e
This might cause a problem of using == instead of equals() for strcmp
if (version == (request.getContext().getWebappVersion())) {
mapRequired = false;
}
When running mod_jk with sticky session, but not using tomcat clustering
... Will adding a new version append the version number to the e
If the Random number generator is "sufficiently random" could we avoid
the digesting? Or should it be an option?
Since java1.6 has SecureRandom - is its current implementation good
enough to avoid all the extra tricks currently being done and we can
just use that as a default? [Or maybe - we s
I checked the svn history of why MD5 (hashing was used) and the picture
is incomplete. (unless someone asks craig since I think he was the author)
But it appears like this ...
Tomcat 3.X use Math.random() and some misc crap to generate its session
id. It had a comment (paraphrased), "not secure
Sorry for the additional noise ... my svn emails are in a different
folder from dev emails. I just noticed ...
svn commit: r1039882 -
/tomcat/trunk/java/org/apache/catalina/session/ManagerBase.java
-Tim
On 11/29/2010 7:40 AM, Tim Funk wrote:
I checked the svn history of why MD5 (hashing
+0.5 - I wonder if in some cases - it may be preferable to use a
property called split which lets the user define the separator which we
can pass to String.split(). [Which OTOH may be more confusing (yet
powerful) since the user is using a regex to split get a list of regex]
-Tim
On 12/24/201
5/2010 2:53 PM, Mark Thomas wrote:
On 25/12/2010 01:49, Tim Funk wrote:
+0.5 - I wonder if in some cases - it may be preferable to use a
property called split which lets the user define the separator which we
can pass to String.split(). [Which OTOH may be more confusing (yet
powerful) since the user
> [X] Stable - go ahead and release as 7.0.6 Stable
-Tim
On 1/10/2011 1:54 PM, Mark Thomas wrote:
The proposed Apache Tomcat 7.0.6 release is now available for voting.
It can be obtained from:
http://people.apache.org/~markt/dev/tomcat-7/v7.0.6/
The svn tag is:
http://svn.apache.org/repos/asf/
> [X] Stable
-Tim
On 1/10/2011 12:18 PM, jean-frederic clere wrote:
The candidates binaries are available here:
http://people.apache.org/~jfclere/tomcat-6/v6.0.30/
According to the release process, the 6.0.30 build corresponding to the
tag TOMCAT_6_0_30 is:
[ ] Broken
[ ] Alpha
[ ] Beta
[ ] St
I don't understand what this change other than provide noise. It doesn't
fix anything.
-Tim
On 1/12/2011 7:08 AM, ma...@apache.org wrote:
Author: markt
Date: Wed Jan 12 12:08:16 2011
New Revision: 1058102
URL: http://svn.apache.org/viewvc?rev=1058102&view=rev
Log:
Fix https://issues.apache.o
JspFactory is in package javax.servlet.jsp which is one of the packages
that has * import. If a user tried to create a class called JspFactory
and tried to import it - he'd get a compile error on his import statement.
-Tim
On 1/12/2011 10:04 AM, Mark Thomas wrote:
On 12/01/2011 14:51
Interesting. Nice nugget to know.
-Tim
On 1/12/2011 10:09 AM, Mark Thomas wrote:
On 12/01/2011 15:08, Tim Funk wrote:
JspFactory is in package javax.servlet.jsp which is one of the packages
that has * import. If a user tried to create a class called JspFactory
and tried to import it - he&#
My preference is to remain CTR on the idea the anything of significance
be proposed so there would be consensus to push it into 7.0 or branch to 7.X
-Tim
On 1/18/2011 1:41 PM, Christopher Schultz wrote:
All,
Since Tomcat 7.0.x went stable, does that change the commit policy, or
are we still
+1 - It might be nice to match 127.0.0.[0-9]{1,3} so it is more virtual
machine friendly for those who map localhost to an alternate loopback
address.
-Tim
On 1/18/2011 7:04 PM, Mark Thomas wrote:
On 19/01/2011 00:00, ma...@apache.org wrote:
Author: markt
Date: Wed Jan 19 00:00:59 2011
New R
My bad - I meant +1 to having it for real.
-Tim
On 1/19/2011 1:48 PM, Mark Thomas wrote:
On 19/01/2011 01:18, Tim Funk wrote:
+1 - It might be nice to match 127.0.0.[0-9]{1,3} so it is more virtual
machine friendly for those who map localhost to an alternate loopback
address.
+1 to having
Do you want to limit the try/catch scope to just
serverSocketFactory.acceptSocket since setSocketOptions() can also throw
IOException?
Is there a case where a client can induce an exception while
setSocketOptions() is processing?
-Tim
On 2/21/2011 10:21 AM, Mark Thomas wrote:
The ASF Sona
their problem.
-Tim
On 2/23/2011 5:48 AM, Mark Thomas wrote:
On 21/02/2011 17:33, Tim Funk wrote:
Do you want to limit the try/catch scope to just
serverSocketFactory.acceptSocket since setSocketOptions() can also throw
IOException?
A couple of people have mentioned this and while the
Would it be more "efficient" to do this instead?
-/** Node identifier when in a cluster. Defaults to the empty string. */
-private String jvmRoute = "";
+/** Node identifier when in a cluster. */
+private String jvmRoute = null;
...
public void setJvmRoute(String jvmRoute) {
Would it be worth making this configurable? For example:
boolean ignoreFailedStarts = false;
...
} catch (LifecycleException e) {
if (ignoreFailedStarts) {
// If realm doesn't start can't authenticate against it
iter.remove();
log.error(sm.getString("combinedRealm.rea
+1 - I like the ideas
+0 - [Day job time constraints to help - but I'll look as I can]
-Tim
On 1/5/2010 7:27 PM, Costin Manolache wrote:
Also, I would like to know if other comitters are OK with (temporarily -
i.e. until the 7.0 release vote) including lite in the tomcat7 builds,
so people can
+1 to trunk
-0 for backport to 6.0.x
Is this legal? ${'\'}${1+1}
Shouldn't it be: ${'\\'}${1+1}
Or phrased another way, what should I see on the screen with this:
${1+1}${'\t'}${1+1}
"2 2" or "2\t2"
-Tim
On 1/8/2010 6:36 AM, Mark Thomas wrote:
One area where help would be appreciated is in
So far all looks OK - but due to the version issue - I vote:
[X] Broken
I believe rebuilding with an updated properties (no retag needed) should
fix the issue.
-Tim
On 2/25/2010 6:07 AM, Konstantin Kolinko wrote:
2010/2/24 jean-frederic clere:
The candidates binaries are available here:
h
On 3/8/2010 2:00 PM, Konstantin Kolinko wrote:
I propose to relax our RTC policy and use CTR for the types of changes
listed below:
2010/3/8 Konstantin Kolinko:
1. We already have Commit-Then-Review for any documentation, including
JavaDoc and code comments.
Already decided. No need to vote.
FWIW ... Broken JSP's in production can be a very common issue.
Or worse - folks who misuse fragments (meant for includes) and leave
them in an area which may be allowed to be requested directly.
-Tim
On 3/10/2010 12:54 PM, Mark Thomas wrote:
I did consider that but I was working on the basi
-0.5
It might seem moving to JIRA from Bugzilla would be a better first step.
Then if commit messages follow a naming convention - the commit could
automagically update the JIRA bug with the commit message. This might
fix the tracking desire yet keeping the simplicity of the STATUS file.
But
I can do an Intro to Tomcat 7. I would think the topics include:
- what is tomcat
- What has not changed vs tomcat6 (highlight various features)
- What is new vs tomcat 7
I hope to update the proposal list today or tomorrow.
-Tim
On 3/19/2010 5:03 AM, Mladen Turk wrote:
On 03/19/2010 09:31 AM,
I'm not a fan of the line backgrounds behind both logos.
From a marketing point of view, I would prefer to see the feather
above the cat.
It might be nice to see the cat and feather swapped and without the
lines in background, just using the gradient.
But otherwise conceptually, it would b
congratulations!
-Tim
On 3/31/2010 3:29 PM, Filip Hanik - Dev Lists wrote:
On behalf of the Tomcat committers I am pleased to announce that Keiichi
Fujino (kfujino) has been voted in as a new Tomcat committer.
Please join me in welcoming him.
Doing this would be bad. When serving JSP's (or anything dynamic greater
than the buffer size) - the content length is not sent to the client. So
when the end of the request is sent - there is no signal to the client
to let them know the request is over and they can start a new request
over the
Wait a sec ... not enough coffee. I might have answered a totally
different question (and incorrectly too)
-Tim
On 4/1/2010 6:42 AM, Tim Funk wrote:
Doing this would be bad. When serving JSP's (or anything dynamic greater
than the buffer size) - the content length is not sent to the c
Agree with Remy. If 'connection: close' is sent by the client, then
tomcat needs to close the connection on the end of the response. So
sending the results via chunked encoding is extra overhead which is not
needed.
A good argument to add is chunked encoding would that it gives the
client the
net effect might be good)
-Tim
On 4/1/2010 6:54 AM, Tim Funk wrote:
A good argument to add is chunked encoding would that it gives the
client the best guess on whether the entire response was returned (by
receiving the last full chunk). But that argument is
This seems like an odd thing to make a constant. (Is something else
setting this or was there an intention for this to be set? Especially
since it is used only once in this file)
-Tim
On 4/12/2010 5:35 AM, ma...@apache.org wrote:
Author: markt
Date: Mon Apr 12 09:35:52 2010
New Revision: 9331
I'm feeling stupid at the moment. (Or need more coffee) But why do the
parenthesis make a difference? (Since only addition/subtraction is done
and no multiplication - I can't tell why this fixes it)
-Tim
On 4/26/2010 5:17 AM, ma...@apache.org wrote:
Author: markt
Date: Mon Apr 26 09:17:46 201
Got it
thanks
-Tim
On 4/26/2010 6:36 AM, Mark Thomas wrote:
On 26/04/2010 11:33, Tim Funk wrote:
I'm feeling stupid at the moment. (Or need more coffee) But why do the
parenthesis make a difference? (Since only addition/subtraction is done
and no multiplication - I can't tell why
I like case 2.
In the case of the malformed request. We might not be able to determine
the host or the path (and therefore the appropriate webapp (or host))
In the common case (admin), I would hope that there is an AccessLog at
the EngineLevel. So we might be able to do the following ...
- C
101 - 200 of 235 matches
Mail list logo
