You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Gary Gregory <ga...@gmail.com> on 2011/03/01 18:59:52 UTC
Re: [codec] Error in the Caverphone implemention
The fix is in SVN and scheduled for inclusion in the upcoming Codec 1.5
release.
Thank you,
Gary
On Tue, Mar 1, 2011 at 6:05 AM, Martin Nybo Andersen <tw...@tweek.dk> wrote:
> Hi,
>
> On Mon, 28 Feb 2011, Matt Benson wrote:
>
> Finally, given the fact that we are discussing patches (which have to
>> be granted IP blah blah blah anyway) as well as the fact that you are
>> not subscribed to the developer list, this exchange should be
>> conducted in JIRA.
>>
>> Matt
>>
>
> I wrote the mail to raise a flag. You saw it, which was my intention. I'm
> happy now.
>
> To create a jira account for this minor bug seems to be overkill for me.
> Remember that the harder it is to report a bug, the fewer bugs will be
> reported.
>
> I am not subscribed to this mailing list for the same reason.
>
> What you do with the bug now, is for you to decide.
>
>
>> On Mon, Feb 28, 2011 at 10:04 AM, Gary Gregory <ga...@gmail.com>
>> wrote:
>>
>>> Would you be able to include a unit test patch in your diff file?
>>>
>>
> new Caverphone().encode("mbmb") currently returns "MMP1111111" while it
> should return "MPM1111111".
>
>
>>> Also, it seems that the examples from
>>>
>>> http://en.wikipedia.org/wiki/Caverphone
>>>
>>> do not match our code:
>>>
>>> {"Lee", "L11111"},
>>> {"Thompson", "TMPSN1"},
>>>
>>>
> These Caverphones seems to be version 1. Version 2 which is implemented in
> apache.commons uses 10-letter-codes.
>
> Hm...
>>>
>>> Thank you,
>>> Gary
>>>
>>> On Mon, Feb 28, 2011 at 8:13 AM, Martin Nybo Andersen <tweek@tweek.dk
>>> >wrote:
>>>
>>> Hi,
>>>>
>>>> I've found an error in the Caverphone language codec.
>>>>
>>>> According to the specs at page 2 line 6:
>>>> "If the name ends with mb make it m2".
>>>>
>>>> Apparently is has been interpreted as:
>>>> "If the name _starts_ with mb make it m2".
>>>>
>>>> The attached patch will fix it.
>>>>
>>>> Please CC me, as I'm not subscribed.
>>>>
>>>
>
> Regards,
> Martin Nybo Andersen
--
Thank you,
Gary
http://garygregory.wordpress.com/
http://garygregory.com/
http://people.apache.org/~ggregory/
http://twitter.com/GaryGregory
Re: [codec] Error in the Caverphone implemention
Posted by Henri Yandell <fl...@gmail.com>.
Heh - pretty sure the 2.0/1.0 bit is me :)
Given I'd written a 1.0 impl first, I think I simply didn't want to
lose the code.
If you think there's value in both, then I'm all for it. I think at
the time I was implementing something fun and didn't know if 2.0 was a
replacement for 1.0 or not. I didn't know if anyone would even use it,
so I'm very happy Martin answered that one :)
Hen
On Thu, Mar 3, 2011 at 10:44 PM, Gary Gregory <ga...@gmail.com> wrote:
> I am not a SME in Caverphone land but I do see in the code comments
> that guide through commenting this line in and that line out to toggle
> between both versions. I would never toggle between versions like that
> nor would I recommend it. This lead me to propose the two classes. In
> addition, it could be useful to provide both versions for programs
> that compare various algorithms. I've seen a couple of papers that
> produce all sorts of stats for different codecs. Reading papers
> recently on cologne and cavedphone while looking for test data is
> where I found such stats for example.
>
> Gary
>
> On Mar 4, 2011, at 1:18, Henri Yandell <fl...@gmail.com> wrote:
>
>> Is there a need for 1.0?
>>
>> On Thu, Mar 3, 2011 at 4:38 AM, Gary Gregory <ga...@gmail.com> wrote:
>>> Any though on splitting the class into two subclasses. One for cav
>>> 1.0, the other for 2.0?
>>>
>>> Gary
>>>
>>> On Mar 3, 2011, at 1:52, Henri Yandell <fl...@gmail.com> wrote:
>>>
>>>> Thanks Martin and Gary - sorry for my screwup.
>>>>
>>>> Very happy that you found the Caverphone code however Martin :)
>>>>
>>>> Hen
>>>>
>>>> On Tue, Mar 1, 2011 at 9:59 AM, Gary Gregory <ga...@gmail.com> wrote:
>>>>> The fix is in SVN and scheduled for inclusion in the upcoming Codec 1.5
>>>>> release.
>>>>>
>>>>> Thank you,
>>>>> Gary
>>>>>
>>>>> On Tue, Mar 1, 2011 at 6:05 AM, Martin Nybo Andersen <tw...@tweek.dk> wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> On Mon, 28 Feb 2011, Matt Benson wrote:
>>>>>>
>>>>>> Finally, given the fact that we are discussing patches (which have to
>>>>>>> be granted IP blah blah blah anyway) as well as the fact that you are
>>>>>>> not subscribed to the developer list, this exchange should be
>>>>>>> conducted in JIRA.
>>>>>>>
>>>>>>> Matt
>>>>>>>
>>>>>>
>>>>>> I wrote the mail to raise a flag. You saw it, which was my intention. I'm
>>>>>> happy now.
>>>>>>
>>>>>> To create a jira account for this minor bug seems to be overkill for me.
>>>>>> Remember that the harder it is to report a bug, the fewer bugs will be
>>>>>> reported.
>>>>>>
>>>>>> I am not subscribed to this mailing list for the same reason.
>>>>>>
>>>>>> What you do with the bug now, is for you to decide.
>>>>>>
>>>>>>
>>>>>>> On Mon, Feb 28, 2011 at 10:04 AM, Gary Gregory <ga...@gmail.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Would you be able to include a unit test patch in your diff file?
>>>>>>>>
>>>>>>>
>>>>>> new Caverphone().encode("mbmb") currently returns "MMP1111111" while it
>>>>>> should return "MPM1111111".
>>>>>>
>>>>>>
>>>>>>>> Also, it seems that the examples from
>>>>>>>>
>>>>>>>> http://en.wikipedia.org/wiki/Caverphone
>>>>>>>>
>>>>>>>> do not match our code:
>>>>>>>>
>>>>>>>> {"Lee", "L11111"},
>>>>>>>> {"Thompson", "TMPSN1"},
>>>>>>>>
>>>>>>>>
>>>>>> These Caverphones seems to be version 1. Version 2 which is implemented in
>>>>>> apache.commons uses 10-letter-codes.
>>>>>>
>>>>>> Hm...
>>>>>>>>
>>>>>>>> Thank you,
>>>>>>>> Gary
>>>>>>>>
>>>>>>>> On Mon, Feb 28, 2011 at 8:13 AM, Martin Nybo Andersen <tweek@tweek.dk
>>>>>>>>> wrote:
>>>>>>>>
>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> I've found an error in the Caverphone language codec.
>>>>>>>>>
>>>>>>>>> According to the specs at page 2 line 6:
>>>>>>>>> "If the name ends with mb make it m2".
>>>>>>>>>
>>>>>>>>> Apparently is has been interpreted as:
>>>>>>>>> "If the name _starts_ with mb make it m2".
>>>>>>>>>
>>>>>>>>> The attached patch will fix it.
>>>>>>>>>
>>>>>>>>> Please CC me, as I'm not subscribed.
>>>>>>>>>
>>>>>>>>
>>>>>>
>>>>>> Regards,
>>>>>> Martin Nybo Andersen
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> Thank you,
>>>>> Gary
>>>>>
>>>>> http://garygregory.wordpress.com/
>>>>> http://garygregory.com/
>>>>> http://people.apache.org/~ggregory/
>>>>> http://twitter.com/GaryGregory
>>>>>
>>>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [codec] Error in the Caverphone implemention
Posted by Gary Gregory <ga...@gmail.com>.
I am not a SME in Caverphone land but I do see in the code comments
that guide through commenting this line in and that line out to toggle
between both versions. I would never toggle between versions like that
nor would I recommend it. This lead me to propose the two classes. In
addition, it could be useful to provide both versions for programs
that compare various algorithms. I've seen a couple of papers that
produce all sorts of stats for different codecs. Reading papers
recently on cologne and cavedphone while looking for test data is
where I found such stats for example.
Gary
On Mar 4, 2011, at 1:18, Henri Yandell <fl...@gmail.com> wrote:
> Is there a need for 1.0?
>
> On Thu, Mar 3, 2011 at 4:38 AM, Gary Gregory <ga...@gmail.com> wrote:
>> Any though on splitting the class into two subclasses. One for cav
>> 1.0, the other for 2.0?
>>
>> Gary
>>
>> On Mar 3, 2011, at 1:52, Henri Yandell <fl...@gmail.com> wrote:
>>
>>> Thanks Martin and Gary - sorry for my screwup.
>>>
>>> Very happy that you found the Caverphone code however Martin :)
>>>
>>> Hen
>>>
>>> On Tue, Mar 1, 2011 at 9:59 AM, Gary Gregory <ga...@gmail.com> wrote:
>>>> The fix is in SVN and scheduled for inclusion in the upcoming Codec 1.5
>>>> release.
>>>>
>>>> Thank you,
>>>> Gary
>>>>
>>>> On Tue, Mar 1, 2011 at 6:05 AM, Martin Nybo Andersen <tw...@tweek.dk> wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> On Mon, 28 Feb 2011, Matt Benson wrote:
>>>>>
>>>>> Finally, given the fact that we are discussing patches (which have to
>>>>>> be granted IP blah blah blah anyway) as well as the fact that you are
>>>>>> not subscribed to the developer list, this exchange should be
>>>>>> conducted in JIRA.
>>>>>>
>>>>>> Matt
>>>>>>
>>>>>
>>>>> I wrote the mail to raise a flag. You saw it, which was my intention. I'm
>>>>> happy now.
>>>>>
>>>>> To create a jira account for this minor bug seems to be overkill for me.
>>>>> Remember that the harder it is to report a bug, the fewer bugs will be
>>>>> reported.
>>>>>
>>>>> I am not subscribed to this mailing list for the same reason.
>>>>>
>>>>> What you do with the bug now, is for you to decide.
>>>>>
>>>>>
>>>>>> On Mon, Feb 28, 2011 at 10:04 AM, Gary Gregory <ga...@gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Would you be able to include a unit test patch in your diff file?
>>>>>>>
>>>>>>
>>>>> new Caverphone().encode("mbmb") currently returns "MMP1111111" while it
>>>>> should return "MPM1111111".
>>>>>
>>>>>
>>>>>>> Also, it seems that the examples from
>>>>>>>
>>>>>>> http://en.wikipedia.org/wiki/Caverphone
>>>>>>>
>>>>>>> do not match our code:
>>>>>>>
>>>>>>> {"Lee", "L11111"},
>>>>>>> {"Thompson", "TMPSN1"},
>>>>>>>
>>>>>>>
>>>>> These Caverphones seems to be version 1. Version 2 which is implemented in
>>>>> apache.commons uses 10-letter-codes.
>>>>>
>>>>> Hm...
>>>>>>>
>>>>>>> Thank you,
>>>>>>> Gary
>>>>>>>
>>>>>>> On Mon, Feb 28, 2011 at 8:13 AM, Martin Nybo Andersen <tweek@tweek.dk
>>>>>>>> wrote:
>>>>>>>
>>>>>>> Hi,
>>>>>>>>
>>>>>>>> I've found an error in the Caverphone language codec.
>>>>>>>>
>>>>>>>> According to the specs at page 2 line 6:
>>>>>>>> "If the name ends with mb make it m2".
>>>>>>>>
>>>>>>>> Apparently is has been interpreted as:
>>>>>>>> "If the name _starts_ with mb make it m2".
>>>>>>>>
>>>>>>>> The attached patch will fix it.
>>>>>>>>
>>>>>>>> Please CC me, as I'm not subscribed.
>>>>>>>>
>>>>>>>
>>>>>
>>>>> Regards,
>>>>> Martin Nybo Andersen
>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Thank you,
>>>> Gary
>>>>
>>>> http://garygregory.wordpress.com/
>>>> http://garygregory.com/
>>>> http://people.apache.org/~ggregory/
>>>> http://twitter.com/GaryGregory
>>>>
>>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [codec] Error in the Caverphone implemention
Posted by Henri Yandell <fl...@gmail.com>.
Is there a need for 1.0?
On Thu, Mar 3, 2011 at 4:38 AM, Gary Gregory <ga...@gmail.com> wrote:
> Any though on splitting the class into two subclasses. One for cav
> 1.0, the other for 2.0?
>
> Gary
>
> On Mar 3, 2011, at 1:52, Henri Yandell <fl...@gmail.com> wrote:
>
>> Thanks Martin and Gary - sorry for my screwup.
>>
>> Very happy that you found the Caverphone code however Martin :)
>>
>> Hen
>>
>> On Tue, Mar 1, 2011 at 9:59 AM, Gary Gregory <ga...@gmail.com> wrote:
>>> The fix is in SVN and scheduled for inclusion in the upcoming Codec 1.5
>>> release.
>>>
>>> Thank you,
>>> Gary
>>>
>>> On Tue, Mar 1, 2011 at 6:05 AM, Martin Nybo Andersen <tw...@tweek.dk> wrote:
>>>
>>>> Hi,
>>>>
>>>> On Mon, 28 Feb 2011, Matt Benson wrote:
>>>>
>>>> Finally, given the fact that we are discussing patches (which have to
>>>>> be granted IP blah blah blah anyway) as well as the fact that you are
>>>>> not subscribed to the developer list, this exchange should be
>>>>> conducted in JIRA.
>>>>>
>>>>> Matt
>>>>>
>>>>
>>>> I wrote the mail to raise a flag. You saw it, which was my intention. I'm
>>>> happy now.
>>>>
>>>> To create a jira account for this minor bug seems to be overkill for me.
>>>> Remember that the harder it is to report a bug, the fewer bugs will be
>>>> reported.
>>>>
>>>> I am not subscribed to this mailing list for the same reason.
>>>>
>>>> What you do with the bug now, is for you to decide.
>>>>
>>>>
>>>>> On Mon, Feb 28, 2011 at 10:04 AM, Gary Gregory <ga...@gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Would you be able to include a unit test patch in your diff file?
>>>>>>
>>>>>
>>>> new Caverphone().encode("mbmb") currently returns "MMP1111111" while it
>>>> should return "MPM1111111".
>>>>
>>>>
>>>>>> Also, it seems that the examples from
>>>>>>
>>>>>> http://en.wikipedia.org/wiki/Caverphone
>>>>>>
>>>>>> do not match our code:
>>>>>>
>>>>>> {"Lee", "L11111"},
>>>>>> {"Thompson", "TMPSN1"},
>>>>>>
>>>>>>
>>>> These Caverphones seems to be version 1. Version 2 which is implemented in
>>>> apache.commons uses 10-letter-codes.
>>>>
>>>> Hm...
>>>>>>
>>>>>> Thank you,
>>>>>> Gary
>>>>>>
>>>>>> On Mon, Feb 28, 2011 at 8:13 AM, Martin Nybo Andersen <tweek@tweek.dk
>>>>>>> wrote:
>>>>>>
>>>>>> Hi,
>>>>>>>
>>>>>>> I've found an error in the Caverphone language codec.
>>>>>>>
>>>>>>> According to the specs at page 2 line 6:
>>>>>>> "If the name ends with mb make it m2".
>>>>>>>
>>>>>>> Apparently is has been interpreted as:
>>>>>>> "If the name _starts_ with mb make it m2".
>>>>>>>
>>>>>>> The attached patch will fix it.
>>>>>>>
>>>>>>> Please CC me, as I'm not subscribed.
>>>>>>>
>>>>>>
>>>>
>>>> Regards,
>>>> Martin Nybo Andersen
>>>
>>>
>>>
>>>
>>> --
>>> Thank you,
>>> Gary
>>>
>>> http://garygregory.wordpress.com/
>>> http://garygregory.com/
>>> http://people.apache.org/~ggregory/
>>> http://twitter.com/GaryGregory
>>>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [codec] Error in the Caverphone implemention
Posted by Gary Gregory <ga...@gmail.com>.
Any though on splitting the class into two subclasses. One for cav
1.0, the other for 2.0?
Gary
On Mar 3, 2011, at 1:52, Henri Yandell <fl...@gmail.com> wrote:
> Thanks Martin and Gary - sorry for my screwup.
>
> Very happy that you found the Caverphone code however Martin :)
>
> Hen
>
> On Tue, Mar 1, 2011 at 9:59 AM, Gary Gregory <ga...@gmail.com> wrote:
>> The fix is in SVN and scheduled for inclusion in the upcoming Codec 1.5
>> release.
>>
>> Thank you,
>> Gary
>>
>> On Tue, Mar 1, 2011 at 6:05 AM, Martin Nybo Andersen <tw...@tweek.dk> wrote:
>>
>>> Hi,
>>>
>>> On Mon, 28 Feb 2011, Matt Benson wrote:
>>>
>>> Finally, given the fact that we are discussing patches (which have to
>>>> be granted IP blah blah blah anyway) as well as the fact that you are
>>>> not subscribed to the developer list, this exchange should be
>>>> conducted in JIRA.
>>>>
>>>> Matt
>>>>
>>>
>>> I wrote the mail to raise a flag. You saw it, which was my intention. I'm
>>> happy now.
>>>
>>> To create a jira account for this minor bug seems to be overkill for me.
>>> Remember that the harder it is to report a bug, the fewer bugs will be
>>> reported.
>>>
>>> I am not subscribed to this mailing list for the same reason.
>>>
>>> What you do with the bug now, is for you to decide.
>>>
>>>
>>>> On Mon, Feb 28, 2011 at 10:04 AM, Gary Gregory <ga...@gmail.com>
>>>> wrote:
>>>>
>>>>> Would you be able to include a unit test patch in your diff file?
>>>>>
>>>>
>>> new Caverphone().encode("mbmb") currently returns "MMP1111111" while it
>>> should return "MPM1111111".
>>>
>>>
>>>>> Also, it seems that the examples from
>>>>>
>>>>> http://en.wikipedia.org/wiki/Caverphone
>>>>>
>>>>> do not match our code:
>>>>>
>>>>> {"Lee", "L11111"},
>>>>> {"Thompson", "TMPSN1"},
>>>>>
>>>>>
>>> These Caverphones seems to be version 1. Version 2 which is implemented in
>>> apache.commons uses 10-letter-codes.
>>>
>>> Hm...
>>>>>
>>>>> Thank you,
>>>>> Gary
>>>>>
>>>>> On Mon, Feb 28, 2011 at 8:13 AM, Martin Nybo Andersen <tweek@tweek.dk
>>>>>> wrote:
>>>>>
>>>>> Hi,
>>>>>>
>>>>>> I've found an error in the Caverphone language codec.
>>>>>>
>>>>>> According to the specs at page 2 line 6:
>>>>>> "If the name ends with mb make it m2".
>>>>>>
>>>>>> Apparently is has been interpreted as:
>>>>>> "If the name _starts_ with mb make it m2".
>>>>>>
>>>>>> The attached patch will fix it.
>>>>>>
>>>>>> Please CC me, as I'm not subscribed.
>>>>>>
>>>>>
>>>
>>> Regards,
>>> Martin Nybo Andersen
>>
>>
>>
>>
>> --
>> Thank you,
>> Gary
>>
>> http://garygregory.wordpress.com/
>> http://garygregory.com/
>> http://people.apache.org/~ggregory/
>> http://twitter.com/GaryGregory
>>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org
Re: [codec] Error in the Caverphone implemention
Posted by Henri Yandell <fl...@gmail.com>.
Thanks Martin and Gary - sorry for my screwup.
Very happy that you found the Caverphone code however Martin :)
Hen
On Tue, Mar 1, 2011 at 9:59 AM, Gary Gregory <ga...@gmail.com> wrote:
> The fix is in SVN and scheduled for inclusion in the upcoming Codec 1.5
> release.
>
> Thank you,
> Gary
>
> On Tue, Mar 1, 2011 at 6:05 AM, Martin Nybo Andersen <tw...@tweek.dk> wrote:
>
>> Hi,
>>
>> On Mon, 28 Feb 2011, Matt Benson wrote:
>>
>> Finally, given the fact that we are discussing patches (which have to
>>> be granted IP blah blah blah anyway) as well as the fact that you are
>>> not subscribed to the developer list, this exchange should be
>>> conducted in JIRA.
>>>
>>> Matt
>>>
>>
>> I wrote the mail to raise a flag. You saw it, which was my intention. I'm
>> happy now.
>>
>> To create a jira account for this minor bug seems to be overkill for me.
>> Remember that the harder it is to report a bug, the fewer bugs will be
>> reported.
>>
>> I am not subscribed to this mailing list for the same reason.
>>
>> What you do with the bug now, is for you to decide.
>>
>>
>>> On Mon, Feb 28, 2011 at 10:04 AM, Gary Gregory <ga...@gmail.com>
>>> wrote:
>>>
>>>> Would you be able to include a unit test patch in your diff file?
>>>>
>>>
>> new Caverphone().encode("mbmb") currently returns "MMP1111111" while it
>> should return "MPM1111111".
>>
>>
>>>> Also, it seems that the examples from
>>>>
>>>> http://en.wikipedia.org/wiki/Caverphone
>>>>
>>>> do not match our code:
>>>>
>>>> {"Lee", "L11111"},
>>>> {"Thompson", "TMPSN1"},
>>>>
>>>>
>> These Caverphones seems to be version 1. Version 2 which is implemented in
>> apache.commons uses 10-letter-codes.
>>
>> Hm...
>>>>
>>>> Thank you,
>>>> Gary
>>>>
>>>> On Mon, Feb 28, 2011 at 8:13 AM, Martin Nybo Andersen <tweek@tweek.dk
>>>> >wrote:
>>>>
>>>> Hi,
>>>>>
>>>>> I've found an error in the Caverphone language codec.
>>>>>
>>>>> According to the specs at page 2 line 6:
>>>>> "If the name ends with mb make it m2".
>>>>>
>>>>> Apparently is has been interpreted as:
>>>>> "If the name _starts_ with mb make it m2".
>>>>>
>>>>> The attached patch will fix it.
>>>>>
>>>>> Please CC me, as I'm not subscribed.
>>>>>
>>>>
>>
>> Regards,
>> Martin Nybo Andersen
>
>
>
>
> --
> Thank you,
> Gary
>
> http://garygregory.wordpress.com/
> http://garygregory.com/
> http://people.apache.org/~ggregory/
> http://twitter.com/GaryGregory
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org