Xpath grep elements - html

Question

I`m using Scrapy Python to try to grep data from the site.
How I can grep this structure with Xpath?
<div class="foo">
<h3>Need this text_1</h3>
<table class="thesamename">
<tbody>
<tr>
<td class="tmp_year">
45767
</td>
<td class="tmp_outcome">
<b>Win_1</b><br>
<span class="tmp_category">TEST_1</span>
</td>
</tr>
<tr>
<td class="tmp_year">
1232004
</td>
<td class="tmp_outcome">
<b>Win_2</b><br>
<span class="tmp_category">TEST_2</span>
</td>
</tr>
<tr>
<td class="tmp_year">
122004
</td>
<td class="tmp_outcome">
<b>Win_3</b><br>
<span class="tmp_category">TEST_3</span>
</td>
</tr>
</tbody>
<h3>Need this text_2</h3>
<table class="thesamename">
<tbody>
<td class="tmp_year">
234
</td>
<td class="tmp_outcome">
<b>Win_E</b><br>
<span class="tmp_category">TEST_E</span>
</td>
</tr>
<tr>
<td class="tmp_year">
3476
</td>
<td class="tmp_outcome">
<b>Win_C</b><br>
<span class="tmp_category">TEST_C</span>
</td>
</tr>
</tbody>
<h3>Need this text_3</h3>
<table class="thesamename">
<tbody>
<tr>
<td class="tmp_year">
85567
</td>
<td class="tmp_outcome">
<b>Win_T</b><br>
<span class="tmp_category">TEST_T</span>
</td>
</tr>
<tr>
<td class="tmp_year">
435656
</td>
<td class="tmp_outcome">
<b>Win_A</b><br>
<span class="tmp_category">TEST_A</span>
</td>
</tr>
<tr>
<td class="tmp_year">
980
</td>
<td class="tmp_outcome">
<b>Win_Z</b><br>
<span class="tmp_category">TEST_Z</span>
</td>
</tr>
</tbody>
I would like to have output with this structure:
"Section": {
Need this text_1 :
[45767 : Win_1 : TEST_1]
[1232004 : Win_2 : TEST_2]
[122004: Win_3 : TEST_3]
,
Need this text_2:
[234 : Win_E : TEST_E]
[3476 : Win_C : TEST_C]
,
Need this text_3:
[85567 : Win_T : TEST_T]
[435656 : Win_A : TEST_A]
[980: Win_Z : TEST_Z]
}
How can I create the proper xpath select to take this structure?
I can take separately all "h3" , all "a" then all tags with class but how I can match?

Answer 1

GREP YOU SAY?! LOL Well, You would be entirely wron to name it so but for the sake ofkeeping the jargon cleanfor understanding your just parsing/extracting.... So new to scrapy? or web dev sideof things? No matter... Theres no way I couldexpect to teach you in one answer here how to xpth/regex like a pro... only wayis for you to keep at but I throw in my input.
First of all, xpath is amazingly usefull wen it comes to websites that are necessarily build to stadard, which doesnt make them bad per say but in the html snipet you gave... its structured all right soo.. Id recommend css extract .. THESE ARE THE VALUES...
year = response.css('td.tmp_year a::text').extract()
outcome = response.css('td.tmp_outcome b::text').extract()
category= response.css('span.tmp_category::text').extract()
PRO-TIP: For what ever case you deem it neccesary, you can save a web page asan HTML file and use scrapy shell by referencing the direct file path to it... So I save you html snippet to a file on my desktop then ran...
scrapy shell file:///home/scriptso/Desktop/letsGREPlol.html
ANYWAYS... as far as xpath... since you asked lol... cake. lets compare the xpath with the cssand tell me you can see... it? lol
response.css('td.tmp_outcome b::text').extract()
so is a td tag....and the class name is tmp_outcome, thn the next node is a bold tag... of which where the text is thusly declaring it as text with the ::text
response.xpath('//td[#class="tmp_outcome"]/b/text()').extract()
So xpath is basically saying we star with a patter inthe entire site of the td tag... and class= tmp_outcome, then the bold, then in xpath to declare type /text() is for text.... /#href is for.. yeah you guessedit