Thread: PATCH: improve sorting in the Data Editor

PATCH: improve sorting in the Data Editor

From
"J.F. Oster"
Date:
Hello,

Here is a small improvement about sorting in the Data editor.

Problem:
If a table has a PK or OIDs, it gets sorted by PK/OID by default.
When a user wants to sort by any other column, he has first to click
"Remove sort" in the context menu, and only then click "Sort
ascending"(descending) on column of his choice.

If he doesn't remove the default sort, he will get no effect and
probably be puzzled for the first time until guesses that he makes
  ORDER BY column_pk, column_of_choice
which is totally useless.
The need to "Remove sort" each time is annoying.

Solution:
When a user chooses to sort by some column, we must discard the
default sort order.

Please see the attached patch.
Thanks.

--
Best regards,
 J.F.

Attachment

Re: PATCH: improve sorting in the Data Editor

From
Ashesh Vashi
Date:
Sanket,

Can you please review it?


--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company


http://www.linkedin.com/in/asheshvashi


On Sun, Aug 2, 2015 at 3:49 PM, J.F. Oster <jinfroster@mail.ru> wrote:
Hello,

Here is a small improvement about sorting in the Data editor.

Problem:
If a table has a PK or OIDs, it gets sorted by PK/OID by default.
When a user wants to sort by any other column, he has first to click
"Remove sort" in the context menu, and only then click "Sort
ascending"(descending) on column of his choice.

If he doesn't remove the default sort, he will get no effect and
probably be puzzled for the first time until guesses that he makes
  ORDER BY column_pk, column_of_choice
which is totally useless.
The need to "Remove sort" each time is annoying.

Solution:
When a user chooses to sort by some column, we must discard the
default sort order.

Please see the attached patch.
Thanks.

--
Best regards,
 J.F.


--
Sent via pgadmin-hackers mailing list (pgadmin-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgadmin-hackers


Re: PATCH: improve sorting in the Data Editor

From
Sanket Mehta
Date:
Sure Ashesh.

Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 10:49 AM, Ashesh Vashi <ashesh.vashi@enterprisedb.com> wrote:
Sanket,

Can you please review it?


--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company


http://www.linkedin.com/in/asheshvashi


On Sun, Aug 2, 2015 at 3:49 PM, J.F. Oster <jinfroster@mail.ru> wrote:
Hello,

Here is a small improvement about sorting in the Data editor.

Problem:
If a table has a PK or OIDs, it gets sorted by PK/OID by default.
When a user wants to sort by any other column, he has first to click
"Remove sort" in the context menu, and only then click "Sort
ascending"(descending) on column of his choice.

If he doesn't remove the default sort, he will get no effect and
probably be puzzled for the first time until guesses that he makes
  ORDER BY column_pk, column_of_choice
which is totally useless.
The need to "Remove sort" each time is annoying.

Solution:
When a user chooses to sort by some column, we must discard the
default sort order.

Please see the attached patch.
Thanks.

--
Best regards,
 J.F.


--
Sent via pgadmin-hackers mailing list (pgadmin-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgadmin-hackers



Re: PATCH: improve sorting in the Data Editor

From
Sanket Mehta
Date:
Hi,

I have reviewed the patch and below is analysis.

- Behaviour of sorting in data editor before applying the patch is as below:

By default table is sorted by PK/OID in ascending order.
when user select any other column to sort the table, that column gets the second priority for sorting the table like i.e.order by col_pk, col_column1 where col_pk is column having PK and col_column1 is any other ordinary column.

- below is the behaviour I have found after applying the patch:

By default table is sorted by PK/OID in ascending order.
When user selects any other column default sorting is removed and table is getting sorted by newly selected column.
But after this if user wants to sort the table by PK or any other column it will have the same behaviour like it has before applying the patch. i.e. order by col_column1, col_PK where col_column1 and col_PK are same as explained above. It will not sort the table by newly selected column(col_PK) because sorting of older column(col_column1) is still in effect and has the higher priority.

This will also confuse the user.

So according to me current behaviour of sorting of table data in pgadmin is acceptable.

Dave, 
Do you have any other thoughts on it?


Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:
Sure Ashesh.

Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 10:49 AM, Ashesh Vashi <ashesh.vashi@enterprisedb.com> wrote:
Sanket,

Can you please review it?


--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company


http://www.linkedin.com/in/asheshvashi


On Sun, Aug 2, 2015 at 3:49 PM, J.F. Oster <jinfroster@mail.ru> wrote:
Hello,

Here is a small improvement about sorting in the Data editor.

Problem:
If a table has a PK or OIDs, it gets sorted by PK/OID by default.
When a user wants to sort by any other column, he has first to click
"Remove sort" in the context menu, and only then click "Sort
ascending"(descending) on column of his choice.

If he doesn't remove the default sort, he will get no effect and
probably be puzzled for the first time until guesses that he makes
  ORDER BY column_pk, column_of_choice
which is totally useless.
The need to "Remove sort" each time is annoying.

Solution:
When a user chooses to sort by some column, we must discard the
default sort order.

Please see the attached patch.
Thanks.

--
Best regards,
 J.F.


--
Sent via pgadmin-hackers mailing list (pgadmin-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgadmin-hackers




Re: PATCH: improve sorting in the Data Editor

From
"J.F. Oster"
Date:
<p>Hello Sanket,<p><br /><p>The behaviour you described after applying the patch is correct in my opinion.<p>That is a
usefulfeature to append further sorting columns after previous ones - the patch has no intent to remove this feature at
all.<p><br/><p>For example, "ORDER BY employee_name, employee_id" may be quite reasonable.<p>Whereas "ORDER BY
employee_id,employee_name" is *always* meaningless.<p><br /><p>Sure, the user could define that meaningless search
orderhimself manually, but that's his fault.<p>Our duty is to avoid this situation caused by default sort order.<p><br
/><p><br/><p><br /><p>Wednesday, August 5, 2015, 8:20:51 PM, you wrote:<p><br /><div><table border="0" cellpadding="1"
cellspacing="2"><trvalign="top"><td style="background-color: #0000ff;" width="12"><p><span
class="rvts6">></span></td><tdstyle="background-color: #ffffff;" width="1125"><p><span
class="rvts7">Hi,</span><p><br/><p><span class="rvts7">I have reviewed the patch and below is analysis.</span><p><br
/><p><spanclass="rvts7">- Behaviour of sorting in data editor before applying the patch is as below:</span><p><br
/><p><spanclass="rvts7">By default table is sorted by PK/OID in ascending order.</span><p><span class="rvts7">when user
selectany other column to sort the table, that column gets the second priority for sorting the table like i.e.order by
col_pk,col_column1 where col_pk is column having PK and col_column1 is any other ordinary column.</span><p><br
/><p><spanclass="rvts7">- below is the behaviour I have found after applying the patch:</span><p><br /><p><span
class="rvts7">Bydefault table is sorted by PK/OID in ascending order.</span><p><span class="rvts7">When user selects
anyother column default sorting is removed and table is getting sorted by newly selected column.</span><p><span
class="rvts7">Butafter this if user wants to sort the table by PK or any other column it will have the same behaviour
likeit has before applying the patch. i.e. order by col_column1, col_PK where col_column1 and col_PK are same as
explainedabove. It will not sort the table by newly selected column(col_PK) because sorting of older
column(col_column1)is still in effect and has the higher priority.</span><p><br /><p><span class="rvts7">This will also
confusethe user.</span><p><br /><p><span class="rvts7">So according to me current behaviour of sorting of table data in
pgadminis acceptable.</span><p><br /><p><span class="rvts7">Dave, </span><p><span class="rvts7">Do you have any other
thoughtson it?</span><p><br /><p><br /><p><br /><p><span class="rvts7">Regards,</span><p><span class="rvts7">Sanket
Mehta</span><p><spanclass="rvts7">Sr Software engineer</span><p><span class="rvts7">Enterprisedb</span><p><br /><p><br
/><p><spanclass="rvts7">On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta <</span><a class="rvts8"
href="mailto:sanket.mehta@enterprisedb.com">sanket.mehta@enterprisedb.com</a><spanclass="rvts7">>
wrote:</span><p><br/><p><span class="rvts7">Sure Ashesh.</span><p><br /><p><br /><p><span
class="rvts7">Regards,</span><p><spanclass="rvts7">Sanket Mehta</span><p><span class="rvts7">Sr Software
engineer</span><p><spanclass="rvts7">Enterprisedb</span><p><br /><p><br /><p><span class="rvts7">On Mon, Aug 3, 2015 at
10:49AM, Ashesh Vashi <</span><a class="rvts8"
href="mailto:ashesh.vashi@enterprisedb.com">ashesh.vashi@enterprisedb.com</a><spanclass="rvts7">>
wrote:</span><p><br/><p><span class="rvts7">Sanket,</span><p><br /><p><span class="rvts7">Can you please review
it?</span><p><br/><p><br /><p><br /><p><span class="rvts9">--</span><p><span class="rvts9">Thanks &
Regards,</span><p><br/><p><span class="rvts9">Ashesh Vashi</span><p><span class="rvts9">EnterpriseDB INDIA: </span><a
class="rvts10"href="http://www.enterprisedb.com">Enterprise PostgreSQL Company</a><p><br /><p><a class="rvts11"
href="http://www.linkedin.com/in/asheshvashi">http://www.linkedin.com/in/asheshvashi</a><p><br/><p><br /><p><span
class="rvts7">OnSun, Aug 2, 2015 at 3:49 PM, J.F. Oster <</span><a class="rvts8"
href="mailto:jinfroster@mail.ru">jinfroster@mail.ru</a><spanclass="rvts7">> wrote:</span><p><span
class="rvts7">Hello,</span><p><br/><p><span class="rvts7">Here is a small improvement about sorting in the Data
editor.</span><p><br/><p><span class="rvts7">Problem:</span><p><span class="rvts7">If a table has a PK or OIDs, it gets
sortedby PK/OID by default.</span><p><span class="rvts7">When a user wants to sort by any other column, he has first to
click</span><p><spanclass="rvts7">"Remove sort" in the context menu, and only then click "Sort</span><p><span
class="rvts7">ascending"(descending)on column of his choice.</span><p><br /><p><span class="rvts7">If he doesn't remove
thedefault sort, he will get no effect and</span><p><span class="rvts7">probably be puzzled for the first time until
guessesthat he makes</span><p><span class="rvts7">  ORDER BY column_pk, column_of_choice</span><p><span
class="rvts7">whichis totally useless.</span><p><span class="rvts7">The need to "Remove sort" each time is
annoying.</span><p><br/><p><span class="rvts7">Solution:</span><p><span class="rvts7">When a user chooses to sort by
somecolumn, we must discard the</span><p><span class="rvts7">default sort order.</span><p><br /><p><span
class="rvts7">Pleasesee the attached patch.</span><p><span class="rvts7">Thanks.</span><p><br /><p><span
class="rvts12">--</span><p><spanclass="rvts12">Best regards,</span><p><span class="rvts12"> J.F.</span><p><br
/><p><spanclass="rvts12">--</span><p><span class="rvts12">Sent via pgadmin-hackers mailing list (</span><a
class="rvts8"href="mailto:pgadmin-hackers@postgresql.org">pgadmin-hackers@postgresql.org</a><span
class="rvts12">)</span><p><spanclass="rvts12">To make changes to your subscription:</span><p><a class="rvts8"
href="http://www.postgresql.org/mailpref/pgadmin-hackers">http://www.postgresql.org/mailpref/pgadmin-hackers</a><p><br
/></td></tr></table></div><p><br/><p><br /><p><br /><p><br /><p><span class="rvts13">-- </span><p><span
class="rvts13">Bestregards,</span><p><span class="rvts13"> J.F.</span> 

Re: PATCH: improve sorting in the Data Editor

From
Dave Page
Date:


On Wed, Aug 5, 2015 at 6:20 PM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:
Hi,

I have reviewed the patch and below is analysis.

- Behaviour of sorting in data editor before applying the patch is as below:

By default table is sorted by PK/OID in ascending order.
when user select any other column to sort the table, that column gets the second priority for sorting the table like i.e.order by col_pk, col_column1 where col_pk is column having PK and col_column1 is any other ordinary column.

- below is the behaviour I have found after applying the patch:

By default table is sorted by PK/OID in ascending order.
When user selects any other column default sorting is removed and table is getting sorted by newly selected column.
But after this if user wants to sort the table by PK or any other column it will have the same behaviour like it has before applying the patch. i.e. order by col_column1, col_PK where col_column1 and col_PK are same as explained above. It will not sort the table by newly selected column(col_PK) because sorting of older column(col_column1) is still in effect and has the higher priority.

This will also confuse the user.

So according to me current behaviour of sorting of table data in pgadmin is acceptable.

Dave, 
Do you have any other thoughts on it?

I think J.F's patch makes sense. It is a little more complex logically speaking, but I also think it's more likely what the user would expect to happen.
 


Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:
Sure Ashesh.

Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 10:49 AM, Ashesh Vashi <ashesh.vashi@enterprisedb.com> wrote:
Sanket,

Can you please review it?


--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company


http://www.linkedin.com/in/asheshvashi


On Sun, Aug 2, 2015 at 3:49 PM, J.F. Oster <jinfroster@mail.ru> wrote:
Hello,

Here is a small improvement about sorting in the Data editor.

Problem:
If a table has a PK or OIDs, it gets sorted by PK/OID by default.
When a user wants to sort by any other column, he has first to click
"Remove sort" in the context menu, and only then click "Sort
ascending"(descending) on column of his choice.

If he doesn't remove the default sort, he will get no effect and
probably be puzzled for the first time until guesses that he makes
  ORDER BY column_pk, column_of_choice
which is totally useless.
The need to "Remove sort" each time is annoying.

Solution:
When a user chooses to sort by some column, we must discard the
default sort order.

Please see the attached patch.
Thanks.

--
Best regards,
 J.F.


--
Sent via pgadmin-hackers mailing list (pgadmin-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgadmin-hackers







--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Re: PATCH: improve sorting in the Data Editor

From
Sanket Mehta
Date:
Hi,

In this case patch looks good to me.

Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Thu, Aug 6, 2015 at 2:22 PM, Dave Page <dpage@pgadmin.org> wrote:


On Wed, Aug 5, 2015 at 6:20 PM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:
Hi,

I have reviewed the patch and below is analysis.

- Behaviour of sorting in data editor before applying the patch is as below:

By default table is sorted by PK/OID in ascending order.
when user select any other column to sort the table, that column gets the second priority for sorting the table like i.e.order by col_pk, col_column1 where col_pk is column having PK and col_column1 is any other ordinary column.

- below is the behaviour I have found after applying the patch:

By default table is sorted by PK/OID in ascending order.
When user selects any other column default sorting is removed and table is getting sorted by newly selected column.
But after this if user wants to sort the table by PK or any other column it will have the same behaviour like it has before applying the patch. i.e. order by col_column1, col_PK where col_column1 and col_PK are same as explained above. It will not sort the table by newly selected column(col_PK) because sorting of older column(col_column1) is still in effect and has the higher priority.

This will also confuse the user.

So according to me current behaviour of sorting of table data in pgadmin is acceptable.

Dave, 
Do you have any other thoughts on it?

I think J.F's patch makes sense. It is a little more complex logically speaking, but I also think it's more likely what the user would expect to happen.
 


Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:
Sure Ashesh.

Regards,
Sanket Mehta
Sr Software engineer
Enterprisedb

On Mon, Aug 3, 2015 at 10:49 AM, Ashesh Vashi <ashesh.vashi@enterprisedb.com> wrote:
Sanket,

Can you please review it?


--

Thanks & Regards,

Ashesh Vashi
EnterpriseDB INDIA: Enterprise PostgreSQL Company


http://www.linkedin.com/in/asheshvashi


On Sun, Aug 2, 2015 at 3:49 PM, J.F. Oster <jinfroster@mail.ru> wrote:
Hello,

Here is a small improvement about sorting in the Data editor.

Problem:
If a table has a PK or OIDs, it gets sorted by PK/OID by default.
When a user wants to sort by any other column, he has first to click
"Remove sort" in the context menu, and only then click "Sort
ascending"(descending) on column of his choice.

If he doesn't remove the default sort, he will get no effect and
probably be puzzled for the first time until guesses that he makes
  ORDER BY column_pk, column_of_choice
which is totally useless.
The need to "Remove sort" each time is annoying.

Solution:
When a user chooses to sort by some column, we must discard the
default sort order.

Please see the attached patch.
Thanks.

--
Best regards,
 J.F.


--
Sent via pgadmin-hackers mailing list (pgadmin-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgadmin-hackers







--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Re: PATCH: improve sorting in the Data Editor

From
"J.F. Oster"
Date:
<p><br /><p>Ashesh, will you commit this, please?<p><br /><p><br /><p>Thursday, August 6, 2015, 12:16:37 PM, you
wrote:<p><br/><div><table border="0" cellpadding="1" cellspacing="2"><tr valign="top"><td style="background-color:
#0000ff;"width="12"><p><span class="rvts6">></span></td><td style="background-color: #ffffff;" width="1125"><p><span
class="rvts7">Hi,</span><p><br/><p><span class="rvts7">In this case patch looks good to me.</span><p><br /><p><br
/><p><spanclass="rvts7">Regards,</span><p><span class="rvts7">Sanket Mehta</span><p><span class="rvts7">Sr Software
engineer</span><p><spanclass="rvts7">Enterprisedb</span><p><br /><p><br /><p><span class="rvts7">On Thu, Aug 6, 2015 at
2:22PM, Dave Page <</span><a class="rvts8" href="mailto:dpage@pgadmin.org">dpage@pgadmin.org</a><span
class="rvts7">>wrote:</span><p><br /><p><br /><p><br /><p><br /><p><br /><p><span class="rvts7">On Wed, Aug 5, 2015
at6:20 PM, Sanket Mehta <</span><a class="rvts8"
href="mailto:sanket.mehta@enterprisedb.com">sanket.mehta@enterprisedb.com</a><spanclass="rvts7">>
wrote:</span><p><br/><p><span class="rvts7">Hi,</span><p><br /><p><span class="rvts7">I have reviewed the patch and
belowis analysis.</span><p><br /><p><span class="rvts7">- Behaviour of sorting in data editor before applying the patch
isas below:</span><p><br /><p><span class="rvts7">By default table is sorted by PK/OID in ascending
order.</span><p><spanclass="rvts7">when user select any other column to sort the table, that column gets the second
priorityfor sorting the table like i.e.order by col_pk, col_column1 where col_pk is column having PK and col_column1 is
anyother ordinary column.</span><p><br /><p><span class="rvts7">- below is the behaviour I have found after applying
thepatch:</span><p><br /><p><span class="rvts7">By default table is sorted by PK/OID in ascending order.</span><p><span
class="rvts7">Whenuser selects any other column default sorting is removed and table is getting sorted by newly
selectedcolumn.</span><p><span class="rvts7">But after this if user wants to sort the table by PK or any other column
itwill have the same behaviour like it has before applying the patch. i.e. order by col_column1, col_PK where
col_column1and col_PK are same as explained above. It will not sort the table by newly selected column(col_PK) because
sortingof older column(col_column1) is still in effect and has the higher priority.</span><p><br /><p><span
class="rvts7">Thiswill also confuse the user.</span><p><br /><p><span class="rvts7">So according to me current
behaviourof sorting of table data in pgadmin is acceptable.</span><p><br /><p><span class="rvts7">Dave, </span><p><span
class="rvts7">Doyou have any other thoughts on it?</span><p><br /><p><span class="rvts7">I think J.F's patch makes
sense.It is a little more complex logically speaking, but I also think it's more likely what the user would expect to
happen.</span><p><spanclass="rvts7"> </span><p><br /><p><br /><p><br /><p><span class="rvts7">Regards,</span><p><span
class="rvts7">SanketMehta</span><p><span class="rvts7">Sr Software engineer</span><p><span
class="rvts7">Enterprisedb</span><p><br/><p><br /><p><span class="rvts7">On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta
<</span><aclass="rvts8" href="mailto:sanket.mehta@enterprisedb.com">sanket.mehta@enterprisedb.com</a><span
class="rvts7">>wrote:</span><p><br /><p><span class="rvts7">Sure Ashesh.</span><p><br /><p><br /><p><span
class="rvts7">Regards,</span><p><spanclass="rvts7">Sanket Mehta</span><p><span class="rvts7">Sr Software
engineer</span><p><spanclass="rvts7">Enterprisedb</span><p><br /><p><br /><p><span class="rvts7">On Mon, Aug 3, 2015 at
10:49AM, Ashesh Vashi <</span><a class="rvts8"
href="mailto:ashesh.vashi@enterprisedb.com">ashesh.vashi@enterprisedb.com</a><spanclass="rvts7">>
wrote:</span><p><br/><p><span class="rvts7">Sanket,</span><p><br /><p><span class="rvts7">Can you please review
it?</span><p><br/><p><br /><p><br /><p><span class="rvts9">--</span><p><span class="rvts9">Thanks &
Regards,</span><p><br/><p><span class="rvts9">Ashesh Vashi</span><p><span class="rvts9">EnterpriseDB INDIA: </span><a
class="rvts10"href="http://www.enterprisedb.com">Enterprise PostgreSQL Company</a><p><br /><p><a class="rvts11"
href="http://www.linkedin.com/in/asheshvashi">http://www.linkedin.com/in/asheshvashi</a><p><br/><p><br /><p><span
class="rvts7">OnSun, Aug 2, 2015 at 3:49 PM, J.F. Oster <</span><a class="rvts8"
href="mailto:jinfroster@mail.ru">jinfroster@mail.ru</a><spanclass="rvts7">> wrote:</span><p><span
class="rvts7">Hello,</span><p><br/><p><span class="rvts7">Here is a small improvement about sorting in the Data
editor.</span><p><br/><p><span class="rvts7">Problem:</span><p><span class="rvts7">If a table has a PK or OIDs, it gets
sortedby PK/OID by default.</span><p><span class="rvts7">When a user wants to sort by any other column, he has first to
click</span><p><spanclass="rvts7">"Remove sort" in the context menu, and only then click "Sort</span><p><span
class="rvts7">ascending"(descending)on column of his choice.</span><p><br /><p><span class="rvts7">If he doesn't remove
thedefault sort, he will get no effect and</span><p><span class="rvts7">probably be puzzled for the first time until
guessesthat he makes</span><p><span class="rvts7">  ORDER BY column_pk, column_of_choice</span><p><span
class="rvts7">whichis totally useless.</span><p><span class="rvts7">The need to "Remove sort" each time is
annoying.</span><p><br/><p><span class="rvts7">Solution:</span><p><span class="rvts7">When a user chooses to sort by
somecolumn, we must discard the</span><p><span class="rvts7">default sort order.</span><p><br /><p><span
class="rvts7">Pleasesee the attached patch.</span><p><span class="rvts7">Thanks.</span><p><br /><p><span
class="rvts12">--</span><p><spanclass="rvts12">Best regards,</span><p><span class="rvts12"> J.F.</span><p><br
/><p><spanclass="rvts12">--</span><p><span class="rvts12">Sent via pgadmin-hackers mailing list (</span><a
class="rvts8"href="mailto:pgadmin-hackers@postgresql.org">pgadmin-hackers@postgresql.org</a><span
class="rvts12">)</span><p><spanclass="rvts12">To make changes to your subscription:</span><p><a class="rvts8"
href="http://www.postgresql.org/mailpref/pgadmin-hackers">http://www.postgresql.org/mailpref/pgadmin-hackers</a><p><br
/><p><br/><p><br /><p><br /><p><br /><p><span class="rvts12">-- </span><p><br /><p><span class="rvts12">Dave
Page</span><p><spanclass="rvts12">Blog: </span><a class="rvts8"
href="http://pgsnake.blogspot.com">http://pgsnake.blogspot.com</a><p><spanclass="rvts12">Twitter: @pgsnake</span><p><br
/><p><spanclass="rvts12">EnterpriseDB UK: </span><a class="rvts8"
href="http://www.enterprisedb.com">http://www.enterprisedb.com</a><p><spanclass="rvts12">The Enterprise PostgreSQL
Company</span></td></tr></table></div><p><br/><p><br /><p><br /><p><br /><p><span class="rvts13">-- </span><p><span
class="rvts13">Bestregards,</span><p><span class="rvts13"> J.F.</span> 

Re: PATCH: improve sorting in the Data Editor

From
"J.F. Oster"
Date:
<p><br /><p>Hi, Dave!<p><br /><p>I see bugreports and patches for PgAdmin III are set aside.<p>What's the current state
ofthe project?<p>If it isn't going to develop further - that should be told to community.<p><br /><p>Seeing one's
effortsbeing ignored simply kills enthusiasm :(<p><br /><p><br /><p>Wednesday, August 12, 2015, 7:06:04 PM, J.F. Oster
wrote:<p><br/><div><table border="0" cellpadding="1" cellspacing="2"><tr valign="top"><td style="background-color:
#0000ff;"width="7"><p><span class="rvts6">></span></td><td style="background-color: #ffffff;" width="1130"><p><br
/><p><spanclass="rvts7">Ashesh, will you commit this, please?</span><p><br /><p><br /><p><span class="rvts7">Thursday,
August6, 2015, 12:16:37 PM, you wrote:</span><p><br /><div><table border="0" cellpadding="1" cellspacing="2"><tr
valign="top"><tdstyle="background-color: #0000ff;" width="23"><p><span class="rvts8">></span></td><td
style="background-color:#ffffff;" width="1101"><p><span class="rvts9">Hi,</span><p><br /><p><span class="rvts9">In this
casepatch looks good to me.</span><p><br /><p><br /><p><span class="rvts9">Regards,</span><p><span class="rvts9">Sanket
Mehta</span><p><spanclass="rvts9">Sr Software engineer</span><p><span class="rvts9">Enterprisedb</span><p><br /><p><br
/><p><spanclass="rvts9">On Thu, Aug 6, 2015 at 2:22 PM, Dave Page <</span><a class="rvts10"
href="mailto:dpage@pgadmin.org">dpage@pgadmin.org</a><spanclass="rvts9">> wrote:</span><p><br /><p><br /><p><br
/><p><br/><p><br /><p><span class="rvts9">On Wed, Aug 5, 2015 at 6:20 PM, Sanket Mehta <</span><a class="rvts10"
href="mailto:sanket.mehta@enterprisedb.com">sanket.mehta@enterprisedb.com</a><spanclass="rvts9">>
wrote:</span><p><br/><p><span class="rvts9">Hi,</span><p><br /><p><span class="rvts9">I have reviewed the patch and
belowis analysis.</span><p><br /><p><span class="rvts9">- Behaviour of sorting in data editor before applying the patch
isas below:</span><p><br /><p><span class="rvts9">By default table is sorted by PK/OID in ascending
order.</span><p><spanclass="rvts9">when user select any other column to sort the table, that column gets the second
priorityfor sorting the table like i.e.order by col_pk, col_column1 where col_pk is column having PK and col_column1 is
anyother ordinary column.</span><p><br /><p><span class="rvts9">- below is the behaviour I have found after applying
thepatch:</span><p><br /><p><span class="rvts9">By default table is sorted by PK/OID in ascending order.</span><p><span
class="rvts9">Whenuser selects any other column default sorting is removed and table is getting sorted by newly
selectedcolumn.</span><p><span class="rvts9">But after this if user wants to sort the table by PK or any other column
itwill have the same behaviour like it has before applying the patch. i.e. order by col_column1, col_PK where
col_column1and col_PK are same as explained above. It will not sort the table by newly selected column(col_PK) because
sortingof older column(col_column1) is still in effect and has the higher priority.</span><p><br /><p><span
class="rvts9">Thiswill also confuse the user.</span><p><br /><p><span class="rvts9">So according to me current
behaviourof sorting of table data in pgadmin is acceptable.</span><p><br /><p><span class="rvts9">Dave, </span><p><span
class="rvts9">Doyou have any other thoughts on it?</span><p><br /><p><span class="rvts9">I think J.F's patch makes
sense.It is a little more complex logically speaking, but I also think it's more likely what the user would expect to
happen.</span><p><spanclass="rvts9"> </span><p><br /><p><br /><p><br /><p><span class="rvts9">Regards,</span><p><span
class="rvts9">SanketMehta</span><p><span class="rvts9">Sr Software engineer</span><p><span
class="rvts9">Enterprisedb</span><p><br/><p><br /><p><span class="rvts9">On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta
<</span><aclass="rvts10" href="mailto:sanket.mehta@enterprisedb.com">sanket.mehta@enterprisedb.com</a><span
class="rvts9">>wrote:</span><p><br /><p><span class="rvts9">Sure Ashesh.</span><p><br /><p><br /><p><span
class="rvts9">Regards,</span><p><spanclass="rvts9">Sanket Mehta</span><p><span class="rvts9">Sr Software
engineer</span><p><spanclass="rvts9">Enterprisedb</span><p><br /><p><br /><p><span class="rvts9">On Mon, Aug 3, 2015 at
10:49AM, Ashesh Vashi <</span><a class="rvts10"
href="mailto:ashesh.vashi@enterprisedb.com">ashesh.vashi@enterprisedb.com</a><spanclass="rvts9">>
wrote:</span><p><br/><p><span class="rvts9">Sanket,</span><p><br /><p><span class="rvts9">Can you please review
it?</span><p><br/><p><br /><p><br /><p><span class="rvts11">--</span><p><span class="rvts11">Thanks &
Regards,</span><p><br/><p><span class="rvts11">Ashesh Vashi</span><p><span class="rvts11">EnterpriseDB INDIA: </span><a
class="rvts12"href="http://www.enterprisedb.com">Enterprise PostgreSQL Company</a><p><br /><p><a class="rvts13"
href="http://www.linkedin.com/in/asheshvashi">http://www.linkedin.com/in/asheshvashi</a><p><br/><p><br /><p><span
class="rvts9">OnSun, Aug 2, 2015 at 3:49 PM, J.F. Oster <</span><a class="rvts10"
href="mailto:jinfroster@mail.ru">jinfroster@mail.ru</a><spanclass="rvts9">> wrote:</span><p><span
class="rvts9">Hello,</span><p><br/><p><span class="rvts9">Here is a small improvement about sorting in the Data
editor.</span><p><br/><p><span class="rvts9">Problem:</span><p><span class="rvts9">If a table has a PK or OIDs, it gets
sortedby PK/OID by default.</span><p><span class="rvts9">When a user wants to sort by any other column, he has first to
click</span><p><spanclass="rvts9">"Remove sort" in the context menu, and only then click "Sort</span><p><span
class="rvts9">ascending"(descending)on column of his choice.</span><p><br /><p><span class="rvts9">If he doesn't remove
thedefault sort, he will get no effect and</span><p><span class="rvts9">probably be puzzled for the first time until
guessesthat he makes</span><p><span class="rvts9">  ORDER BY column_pk, column_of_choice</span><p><span
class="rvts9">whichis totally useless.</span><p><span class="rvts9">The need to "Remove sort" each time is
annoying.</span><p><br/><p><span class="rvts9">Solution:</span><p><span class="rvts9">When a user chooses to sort by
somecolumn, we must discard the</span><p><span class="rvts9">default sort order.</span><p><br /><p><span
class="rvts9">Pleasesee the attached patch.</span><p><span class="rvts9">Thanks.</span><p><br /><p><span
class="rvts14">--</span><p><spanclass="rvts14">Best regards,</span><p><span class="rvts14"> J.F.</span><p><br
/><p><spanclass="rvts14">--</span><p><span class="rvts14">Sent via pgadmin-hackers mailing list (</span><a
class="rvts10"href="mailto:pgadmin-hackers@postgresql.org">pgadmin-hackers@postgresql.org</a><span
class="rvts14">)</span><p><spanclass="rvts14">To make changes to your subscription:</span><p><a class="rvts10"
href="http://www.postgresql.org/mailpref/pgadmin-hackers">http://www.postgresql.org/mailpref/pgadmin-hackers</a><p><br
/><p><br/><p><br /><p><br /><p><br /><p><span class="rvts14">-- </span><p><br /><p><span class="rvts14">Dave
Page</span><p><spanclass="rvts14">Blog: </span><a class="rvts10"
href="http://pgsnake.blogspot.com">http://pgsnake.blogspot.com</a><p><spanclass="rvts14">Twitter: @pgsnake</span><p><br
/><p><spanclass="rvts14">EnterpriseDB UK: </span><a class="rvts10"
href="http://www.enterprisedb.com">http://www.enterprisedb.com</a><p><spanclass="rvts14">The Enterprise PostgreSQL
Company</span></td></tr></table></div><p><br/><p><br /><p><br /><p><br /><p><span class="rvts15">-- </span><p><span
class="rvts15">Bestregards,</span><p><span class="rvts15"> J.F.</span></td></tr></table></div><p><br /><p><br /><p><br
/><p><br/><p><span class="rvts16">-- </span><p><span class="rvts16">Best regards,</span><p><span
class="rvts16"> J.F.</span>

Re: PATCH: improve sorting in the Data Editor

From
Dave Page
Date:
Hi

On Wed, Sep 2, 2015 at 6:31 AM, J.F. Oster <jinfroster@mail.ru> wrote:


Hi, Dave!


I see bugreports and patches for PgAdmin III are set aside.

Not set aside - it's just that the people on my team are busy with other work at the moment. There are other committers though.
 

What's the current state of the project?

If it isn't going to develop further - that should be told to community.

pgAdmin III will be around for at least another year, but EDB staff are concentrating mostly on pgAdmin 4 at the moment. That's been the subject of blogs and talks at various conferences, and you should have seem commits and some discussion here I expect.
 


Seeing one's efforts being ignored simply kills enthusiasm :(



Wednesday, August 12, 2015, 7:06:04 PM, J.F. Oster wrote:


>


Ashesh, will you commit this, please?



Thursday, August 6, 2015, 12:16:37 PM, you wrote:


>

Hi,


In this case patch looks good to me.



Regards,

Sanket Mehta

Sr Software engineer

Enterprisedb



On Thu, Aug 6, 2015 at 2:22 PM, Dave Page <dpage@pgadmin.org> wrote:






On Wed, Aug 5, 2015 at 6:20 PM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:


Hi,


I have reviewed the patch and below is analysis.


- Behaviour of sorting in data editor before applying the patch is as below:


By default table is sorted by PK/OID in ascending order.

when user select any other column to sort the table, that column gets the second priority for sorting the table like i.e.order by col_pk, col_column1 where col_pk is column having PK and col_column1 is any other ordinary column.


- below is the behaviour I have found after applying the patch:


By default table is sorted by PK/OID in ascending order.

When user selects any other column default sorting is removed and table is getting sorted by newly selected column.

But after this if user wants to sort the table by PK or any other column it will have the same behaviour like it has before applying the patch. i.e. order by col_column1, col_PK where col_column1 and col_PK are same as explained above. It will not sort the table by newly selected column(col_PK) because sorting of older column(col_column1) is still in effect and has the higher priority.


This will also confuse the user.


So according to me current behaviour of sorting of table data in pgadmin is acceptable.


Dave, 

Do you have any other thoughts on it?


I think J.F's patch makes sense. It is a little more complex logically speaking, but I also think it's more likely what the user would expect to happen.

 




Regards,

Sanket Mehta

Sr Software engineer

Enterprisedb



On Mon, Aug 3, 2015 at 11:07 AM, Sanket Mehta <sanket.mehta@enterprisedb.com> wrote:


Sure Ashesh.



Regards,

Sanket Mehta

Sr Software engineer

Enterprisedb



On Mon, Aug 3, 2015 at 10:49 AM, Ashesh Vashi <ashesh.vashi@enterprisedb.com> wrote:


Sanket,


Can you please review it?




--

Thanks & Regards,


Ashesh Vashi

EnterpriseDB INDIA: Enterprise PostgreSQL Company


http://www.linkedin.com/in/asheshvashi



On Sun, Aug 2, 2015 at 3:49 PM, J.F. Oster <jinfroster@mail.ru> wrote:

Hello,


Here is a small improvement about sorting in the Data editor.


Problem:

If a table has a PK or OIDs, it gets sorted by PK/OID by default.

When a user wants to sort by any other column, he has first to click

"Remove sort" in the context menu, and only then click "Sort

ascending"(descending) on column of his choice.


If he doesn't remove the default sort, he will get no effect and

probably be puzzled for the first time until guesses that he makes

  ORDER BY column_pk, column_of_choice

which is totally useless.

The need to "Remove sort" each time is annoying.


Solution:

When a user chooses to sort by some column, we must discard the

default sort order.


Please see the attached patch.

Thanks.


--

Best regards,

 J.F.


--

Sent via pgadmin-hackers mailing list (pgadmin-hackers@postgresql.org)

To make changes to your subscription:

http://www.postgresql.org/mailpref/pgadmin-hackers






-- 


Dave Page

Blog: http://pgsnake.blogspot.com

Twitter: @pgsnake


EnterpriseDB UK: http://www.enterprisedb.com

The Enterprise PostgreSQL Company





-- 

Best regards,

 J.F.





-- 

Best regards,

 J.F.




--
Dave Page
Blog: http://pgsnake.blogspot.com
Twitter: @pgsnake

EnterpriseDB UK: http://www.enterprisedb.com
The Enterprise PostgreSQL Company