Re: [PATCH] doc/queries.sgml: add missing comma
On Tue, Aug 30, 2022 at 02:18:42PM -0400, Bruce Momjian wrote: > On Wed, Aug 24, 2022 at 07:58:04PM +0200, Peter Eisentraut wrote: > > On 18.08.22 20:10, Bruce Momjian wrote: > > > > Thus: > > > > Strictly speaking, this process is iteration, but > > > > RECURSIVE > > > > is the terminology chosen by the SQL standards committee." > > > > > > > > Because the above sounds just fine, I'd argue that if one does leave > > > > "not > > > > recursion" it should be set off by a comma. > > > I went with new wording, which should make this even clearer; patch > > > attached. > > > > I think this whole note is a bit misleading, like the SQL people don't know > > what recursion is. The point is that the query is defined recursively. The > > evaluation process is iterative. Those two are not contradictions. > > Okay, makes sense. Here is an updated patch. Patch applied back to PG 10. -- Bruce Momjian https://momjian.us EDB https://enterprisedb.com Indecision is a decision. Inaction is an action. Mark Batterson
Re: [PATCH] doc/queries.sgml: add missing comma
On Wed, Aug 24, 2022 at 07:58:04PM +0200, Peter Eisentraut wrote: > On 18.08.22 20:10, Bruce Momjian wrote: > > > Thus: > > > Strictly speaking, this process is iteration, but > > > RECURSIVE > > > is the terminology chosen by the SQL standards committee." > > > > > > Because the above sounds just fine, I'd argue that if one does leave "not > > > recursion" it should be set off by a comma. > > I went with new wording, which should make this even clearer; patch > > attached. > > I think this whole note is a bit misleading, like the SQL people don't know > what recursion is. The point is that the query is defined recursively. The > evaluation process is iterative. Those two are not contradictions. Okay, makes sense. Here is an updated patch. -- Bruce Momjian https://momjian.us EDB https://enterprisedb.com Indecision is a decision. Inaction is an action. Mark Batterson diff --git a/doc/src/sgml/queries.sgml b/doc/src/sgml/queries.sgml index 1428d99d0f..4c5a83c9cb 100644 --- a/doc/src/sgml/queries.sgml +++ b/doc/src/sgml/queries.sgml @@ -2172,9 +2172,8 @@ SELECT sum(n) FROM t; -Strictly speaking, this process is iteration not recursion, but -RECURSIVE is the terminology chosen by the SQL standards -committee. +While RECURSIVE allows queries to be specified +recursively, internally all queries are evaluated iteratively.
Re: [PATCH] doc/queries.sgml: add missing comma
On 18.08.22 20:10, Bruce Momjian wrote: Thus: Strictly speaking, this process is iteration, but RECURSIVE is the terminology chosen by the SQL standards committee." Because the above sounds just fine, I'd argue that if one does leave "not recursion" it should be set off by a comma. I went with new wording, which should make this even clearer; patch attached. I think this whole note is a bit misleading, like the SQL people don't know what recursion is. The point is that the query is defined recursively. The evaluation process is iterative. Those two are not contradictions.
Re: [PATCH] doc/queries.sgml: add missing comma
On Tue, Oct 19, 2021 at 06:41:54PM -0700, David G. Johnston wrote: > On Tue, Oct 19, 2021 at 2:33 PM Tom Lane wrote: > > Drew DeVault writes: > > Minor grammatical fix. > > Hmm, I'm not sure that reads any better than before. > > > > > - Strictly speaking, this process is iteration not recursion, but > > + Strictly speaking, this process is iteration, not recursion, but > > RECURSIVE is the terminology chosen by the SQL > standards > > ... although I think this text is mine, so naturally I'd think > that. Anyone else have an opinion? > > > > If I read that aloud to myself there is a comma after iteration. > > That said, given that a comma and a "but" later we use the word "RECURSIVE" > the > clarification that the process isn't recursion seems redundant. If one knows > what it means to be "recursive" then they will understand the juxtaposition of > "iteration" and "recursive" just fine. If they do not, I don't think adding > the word "recursion" is going to make much difference. > > Thus: > Strictly speaking, this process is iteration, but RECURSIVE > is the terminology chosen by the SQL standards committee." > > Because the above sounds just fine, I'd argue that if one does leave "not > recursion" it should be set off by a comma. I went with new wording, which should make this even clearer; patch attached. -- Bruce Momjian https://momjian.us EDB https://enterprisedb.com Indecision is a decision. Inaction is an action. Mark Batterson diff --git a/doc/src/sgml/queries.sgml b/doc/src/sgml/queries.sgml index 516fbcbf37..51b752709e 100644 --- a/doc/src/sgml/queries.sgml +++ b/doc/src/sgml/queries.sgml @@ -2172,7 +2172,7 @@ SELECT sum(n) FROM t; -Strictly speaking, this process is iteration not recursion, but +Strictly speaking, this process is iterative, not recursive, but RECURSIVE is the terminology chosen by the SQL standards committee.
Re: [PATCH] doc/queries.sgml: add missing comma
On Tue, Oct 19, 2021 at 06:41:54PM -0700, David G. Johnston wrote: > On Tue, Oct 19, 2021 at 2:33 PM Tom Lane wrote: >> ... although I think this text is mine, so naturally I'd think >> that. Anyone else have an opinion? > > If I read that aloud to myself there is a comma after iteration. Agreed, adding a comma feels more natural. Non-native speaker here, however. -- Michael signature.asc Description: PGP signature
Re: [PATCH] doc/queries.sgml: add missing comma
On Tue, Oct 19, 2021 at 2:33 PM Tom Lane wrote: > Drew DeVault writes: > > Minor grammatical fix. > > Hmm, I'm not sure that reads any better than before. > > > > > -Strictly speaking, this process is iteration not recursion, but > > +Strictly speaking, this process is iteration, not recursion, but > > RECURSIVE is the terminology chosen by the SQL > standards > > ... although I think this text is mine, so naturally I'd think > that. Anyone else have an opinion? > > If I read that aloud to myself there is a comma after iteration. That said, given that a comma and a "but" later we use the word "RECURSIVE" the clarification that the process isn't recursion seems redundant. If one knows what it means to be "recursive" then they will understand the juxtaposition of "iteration" and "recursive" just fine. If they do not, I don't think adding the word "recursion" is going to make much difference. Thus: Strictly speaking, this process is iteration, but RECURSIVE is the terminology chosen by the SQL standards committee." Because the above sounds just fine, I'd argue that if one does leave "not recursion" it should be set off by a comma. David J.
Re: [PATCH] doc/queries.sgml: add missing comma
Drew DeVault writes: > Minor grammatical fix. Hmm, I'm not sure that reads any better than before. > > -Strictly speaking, this process is iteration not recursion, but > +Strictly speaking, this process is iteration, not recursion, but > RECURSIVE is the terminology chosen by the SQL > standards ... although I think this text is mine, so naturally I'd think that. Anyone else have an opinion? regards, tom lane
