sql - one record from each group? -

-

मेरे पास इस तरह एक मेज है: 

  x1 | Y1 | X2 | Y2 | क्षेत्र | रंग 5 | 0 | 5 | 0 | 1 | आर 5 | 0 | 6 | 0 | 2 | जी 5 | 1 | 5 | 0 | 2 | बी 5 | 1 | 5 | 1 | 2 | आर 5 | 2 | 5 | 0 | 3 | जी 5 | 2 | 5 | 1 | 3 | बी

मैं एक एसक्यूएल क्वेरी का निर्माण कैसे कर सकता हूं ताकि परिणामस्वरूप तालिका में एक आर, जी, बी (अधिकतम क्षेत्र वाले प्रत्येक) के एक < Em> इस रिकॉर्ड के निर्देशांक को संरक्षित करें? मैंने MAX (क्षेत्र) और ग्रुप द्वारा रंग की कोशिश की है, लेकिन इससे मुझे आर, जी, बी के लिए अधिकतम क्षेत्र के साथ रिकार्ड मिल जाता है, लेकिन उनके निर्देशांक के साथ मिलकर नहीं।

मैं PostgreSQL नमूना आउटपुट का उपयोग कर रहा हूं (ऊपर के लिए ), कुछ ऐसा होना चाहिए: 

  5 | 1 | 5 | 1 | 2 | आर 5 | 2 | 5 | 0 | 3 | जी 5 | 2 | 5 | 1 | 3 | बी

असल में इसे चौथा, 5 वें और 6 वें रिकॉर्ड को वापस करना चाहिए।

आपका प्रश्न थोड़ा अस्पष्ट है, लेकिन मुझे लगता है कि आप यह चाहते हैं: 

  SELECT T2। * से (चयन रंग, अधिकतम (क्षेत्र) एएस क्षेत्र से तालिका 1 ग्रुप द्वारा रंग) एएस टी 1 तालिका 1 में जुड़ें टी 2 ऑन टी 1.रेआ = टी 2.रेआ और टी 1 कॉरलर = टी 2 कॉरल   
 ध्यान दें कि यह मानता है कि अधिकतम क्षेत्र अद्वितीय है, अन्यथा आपको कई पंक्तियां मिलेंगी। < / P> 
 आपके डेटा के लिए आउटपुट:  
  x1 | X2 | Y1 | Y2 | क्षेत्र | रंग 5 | 1 | 5 | 1 | 2 | आर 5 | 2 | 5 | 0 | 3 | जी 5 | 2 | 5 | 1 | 3 | बी  
 
 यदि आप एक बेहतर जवाब चाहते हैं, तो कृपया उन सभी सवालों को स्पष्ट करें जिन्हें मैंने टिप्पणी के रूप में पोस्ट किया है (उनमें से कुछ छिपे हुए हैं)।

Comments

Post a Comment

Popular posts from this blog

objective c - iPhone and it's wireless area -

-
Is this a way to find other iPhones in the category of wireless areas? So I give an example. I'm staying at the airport somewhere I have an iPhone too. I and the other iPhone owners are playing the same game. Is there any way to warn me and the other player that we are close to each other? Does this work even if we are not on the same wireless network, or there is no wireless? And if there is any way to do this, can you give me some starting point? Like a link where I can learn something like this or maybe this is probably the name of this process, because I do not know how to search about this topic in Google. Thanks in advance! Can be completed with peer-to-peer discovery and connectivity you are describing Game Kit Framework Originally I thought it works only through Bluetooth, but after watching it immediately, it seems that WiFi is also supported. Hope it helps!
Read more

unix - Listing all shared memory segments used by a process on AIX5.3+ -

-
I would like to find all the shared memory segments used by a given process I am particularly interested in detecting shmid That's why I can use it in call (shmctl). On Solaris I will just read / proc / $ PID / map to find that information (field pr_shmid) The content of that file is defined by struct prmap_t in the sys / procfs format. Ax also has a / proc / $ PID / map file. Any project is a primate, but unfortunately it has been projected in pr_shmid field. Any ideas how can I get it on AIX5.3 +? I do not particularly know about AX, but I think that ipcs command is quite standard where SysV IPC is supported, so I want to get the ipcs -m command to give the appropriate code; If you can not find a better method, then there may be an option to parse its output.
Read more

tsql - Correct sql/hql query (aggregate in where clause) -

-
I want to query the query as below but the question is wrong but describes my intentions. select name, date time, data from record, date time = maximum (date time) Update: OK The description of the query intentions is not very good. My bad I want to select the latest record for each person. i like Mickey's answer and quosnoye (and upyoyed miki), but if your needs are similar to me , You should keep in mind some limitations. First and foremost, this works only when you are looking for the latest record or latest record for the same name. If you want the latest record for each person in a set (a record for each person but the latest record for each), then the above solutions are small. Second, and less importantly, if you work with a large dataset, then it can be slow for a long time. So, what's around the work? Whatever I do is to add a small field to mark the "latest" in the table. Then, when I store a record (which is done in the stored procedu...
Read more