Thank you, very interesting.

On Monday, February 9, 2015 at 12:28:41 AM UTC+1, Lvc@ wrote:
>
> Hash indexes can be O(1) in the best case, so I'd say between O(logN) and 
> O(1). However direct access is always faster, even than Hash index when 
> it's O(1).
>
> Lvc@
>
> ᐧ
>
> On 8 February 2015 at 22:12, Riccardo Tasso <[email protected] 
> <javascript:>> wrote:
>
>> Aren't hash indexes O(1)?
>>
>> Riccardo
>> Il 08/feb/2015 21:08 "Luca Garulli" <[email protected] <javascript:>> ha 
>> scritto:
>>
>> Hi Rasmus,
>>> Indexes tend to be O(LogN), while RID is O(1). So it will always be 
>>> faster.
>>>
>>> Lvc@
>>>
>>> ᐧ
>>>
>>> On 8 February 2015 at 11:39, Rasmus Eneman <[email protected] 
>>> <javascript:>> wrote:
>>>
>>>> Luca, bu using slugs in your URLs instead of database IDs you 
>>>> completely remove this problem, as well as making your urls nicer and 
>>>> better for SEO.
>>>>
>>>> A question for people with good experience with OrientDB, what is the 
>>>> performance implications with using slugs over rids if the slugs are 
>>>> indexed?
>>>> Should I take care to keep a slug to rid mapping in for example redis 
>>>> so that queries to OrientDB always uses rids?
>>>>
>>>>
>>>> On Wednesday, January 28, 2015 at 12:43:21 AM UTC+1, JR wrote:
>>>>>
>>>>> Thanks Riccardo,
>>>>>
>>>>> good point! do you know how others are solving this problem? I'm 
>>>>> planning to use OrientDB as my only storage solution.
>>>>>
>>>>> Thanks a lot,
>>>>> Luca
>>>>>
>>>>> On Tuesday, January 27, 2015 at 3:39:54 PM UTC-8, Riccardo Tasso wrote:
>>>>>>
>>>>>> Yes Luca,
>>>>>>
>>>>>> consider also if you use rids in your URLs. For example: 
>>>>>> http://myapp.com/details.jsp?rid=1:1
>>>>>>
>>>>>> In this case if someone links to your URLs it could be a problem.
>>>>>>
>>>>>> Cheers,
>>>>>>    Riccardo
>>>>>> Il 27/gen/2015 21:29 "Luca Rondanini" <[email protected]> ha 
>>>>>> scritto:
>>>>>>
>>>>>>> Hi Andrey,
>>>>>>>
>>>>>>> just to double check, if I use RIDs in my web app everything should 
>>>>>>> be fine. Inserts/updates won't change any RIDs. Problems may raise if, 
>>>>>>> for 
>>>>>>> example, I'm storing RIDs in MySQL as reference. If then, I export and 
>>>>>>> reimport the data in orientDB the RIDs may be changed and the MySQL 
>>>>>>> data 
>>>>>>> won't reference the right rows anymore.
>>>>>>>
>>>>>>> Can you confirm?
>>>>>>>
>>>>>>> Thanks,
>>>>>>> Luca
>>>>>>>
>>>>>>> On Monday, January 19, 2015 at 5:28:16 AM UTC-8, Andrey Lomakin 
>>>>>>> wrote:
>>>>>>>>
>>>>>>>> Hi Bertran,
>>>>>>>> The only issue is that if you do export database to JSON and then 
>>>>>>>> back your RIDs will be changed.
>>>>>>>> But may be we will add option to keep rids on their places. Not so 
>>>>>>>> hard to do. 
>>>>>>>> The only downside of such flag will be presence of holes for rids 
>>>>>>>> of removed records.
>>>>>>>> Each record removal consumes about 10 bytes of disk space (not so 
>>>>>>>> much )) ), but when you do export/import we optimize space is used by 
>>>>>>>> database as result rids may be changed.
>>>>>>>>
>>>>>>>> On Mon, Jan 19, 2015 at 2:48 PM, Bertrand Quenin <
>>>>>>>> [email protected]> wrote:
>>>>>>>>
>>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> I'm currently using OrientDB for a project at work and I have some 
>>>>>>>>> question about record IDs.
>>>>>>>>>
>>>>>>>>> As far as I understood, record IDs are unique and can be used to 
>>>>>>>>> to identify a record in a deterministic way. I was wondering if these 
>>>>>>>>> IDs 
>>>>>>>>> can be used to identify objects through (REST) APIs.
>>>>>>>>>
>>>>>>>>> As any web application, I need to identify objects one way or 
>>>>>>>>> another and I was wondering what the best approach would be:
>>>>>>>>> 1/ Using record IDs provided by orient DB
>>>>>>>>> 2/ Adding a "business" id field to my objects and manage it aside.
>>>>>>>>>
>>>>>>>>> So far, I didn't see any reason why I shouldn't use the record ID 
>>>>>>>>> itself, the only con I see is that RID are strings clusterId#position 
>>>>>>>>> (and 
>>>>>>>>> not plain integers) and it's a bit harder to expose in an URL ...
>>>>>>>>>
>>>>>>>>> Is there any issue I haven't foresee ?
>>>>>>>>>
>>>>>>>>> Thanks for your help
>>>>>>>>>
>>>>>>>>> -- 
>>>>>>>>>
>>>>>>>>> --- 
>>>>>>>>> You received this message because you are subscribed to the Google 
>>>>>>>>> Groups "OrientDB" group.
>>>>>>>>> To unsubscribe from this group and stop receiving emails from it, 
>>>>>>>>> send an email to [email protected].
>>>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> -- 
>>>>>>>> Best regards,
>>>>>>>> Andrey Lomakin.
>>>>>>>>
>>>>>>>>   -- 
>>>>>>>
>>>>>>> --- 
>>>>>>> You received this message because you are subscribed to the Google 
>>>>>>> Groups "OrientDB" group.
>>>>>>> To unsubscribe from this group and stop receiving emails from it, 
>>>>>>> send an email to [email protected].
>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>>
>>>>>>  -- 
>>>>
>>>> --- 
>>>> You received this message because you are subscribed to the Google 
>>>> Groups "OrientDB" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>> an email to [email protected] <javascript:>.
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>>
>>>  -- 
>>>
>>> --- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "OrientDB" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to [email protected] <javascript:>.
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>  -- 
>>
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "OrientDB" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"OrientDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to