From 40fc1eb64fb8da26544bd2e2b4855813ffb2f244 Mon Sep 17 00:00:00 2001
From: msglm <msglm@techchud.xyz>
Date: Wed, 11 Jan 2023 20:49:27 -0600
Subject: cache system actually works now

---
 src/indeedwatcher.nim | 148 +++++++++++++++++++++++++++-----------------------
 1 file changed, 79 insertions(+), 69 deletions(-)

(limited to 'src')

diff --git a/src/indeedwatcher.nim b/src/indeedwatcher.nim
index 1807e93..2e5ff87 100644
--- a/src/indeedwatcher.nim
+++ b/src/indeedwatcher.nim
@@ -4,83 +4,93 @@ import osproc
 import options
 import os
 import strutils
+import parsetoml
+import sequtils
 
-#Feednim
-var test = getRSS("https://rss.indeed.com/rss?q=Linux&l=Arkansas&explvl=mid_level")
-var URL = test.items[5].link
+
+
+#TODO make this create folders and files for this automatically upon first start up
+let config = parsetoml.parseFile(getConfigDir() & "/indeedwatcher/config.toml")
+let feeds = config["feeds"].getElems().mapIt(it.getStr())
+let outdir = config["outdir"].getStr()
+let chromedriverloc = config["chromedriver"].getStr()
+let cachefileloc = getCacheDir() & "/indeedwatcher/listings.cache" 
+var cache = splitLines(readFile(cachefileloc))
 
 #Webdriver
-#TODO put location of chromedriver into config
-let chromedriver = startProcess("/usr/bin/chromedriver")
+let chromedriver = startProcess(chromedriverloc, "", ["--headless", "--disable-gpu"])
 sleep 5000
 echo "connecting"
+#TODO make the port configurable, some users may have something running here
 let driver = newWebDriver("http://localhost:9515")
 let session = driver.createSession()
-session.navigate(URL)
-
-#HTML Parser
-var jobTimes: string
-var salaryGuide: string
-
-let jobTitle = session.findElement(".jobsearch-JobInfoHeader-title").get().getText()
-let employer = session.findElement(".jobsearch-CompanyReview--heading").get().getText()
-
-try:
-    jobTimes = session.findElement(".jobsearch-JobDescriptionSection-sectionItem").get().getText()
-except UnpackDefect:
-    jobTimes = ""
+var counter = 0
+
+proc terminate() {.noconv.} = 
+    echo "\nAcknowledged termination attempt..."
+    echo "Writing Cache..."
+    writeFile(cachefileloc, cache.join("\n"))
+    echo "Closing Session..."
+    session.close()
+    echo "Killing Chromedriver..."
+    terminate(chromedriver)
+    echo "Dying!"
+    quit()
+setControlCHook(terminate)
+
+for feed in feeds:
+    echo "now reading " & feed
+
+    #Feednim
+    var rssFeedReply = getRSS(feed)
+    
+    for entry in rssFeedReply.items:
+        echo entry.link
+        #Logging
+        
+        if counter > 7:
+            echo "resting for 7 seconds ..."
+            sleep 7000
+            counter = 0
+        var URL = entry.link
+        let URLID = entry.link.split('&')[4]
+        echo any(cache, proc (input: string): bool = input.contains(URLID))
+        if not any(cache, proc (input: string): bool = input.contains(URLID)):
+            session.navigate(URL)
+            counter = counter + 1
+        
+            #HTML Parser
+            var salaryInfoAndJobType: string
+            var employer: string
+            let jobTitle = session.findElement(".jobsearch-JobInfoHeader-title").get().getText()
+            
+            try:
+                employer = session.findElement(".jobsearch-InlineCompanyRating-companyHeader").get().getText()
+            except UnpackDefect:
+                salaryInfoAndJobType = "None Listed"
+            
+            try:
+                salaryInfoAndJobType = session.findelement("#salaryInfoAndJobType").get().gettext()
+            except UnpackDefect:
+                salaryInfoAndJobType = "None Listed"
+            
+            let fullDesc = session.findElement("#jobDescriptionText").get().getText()
+            
+            #Job Value Scorer
+            
+            #Output
+            var output = """
+            Title: $1
+            Company: $2
+            Salary Info and Job Type: $3
+            URL : $4
+            Description:
+            $5
+            """ % [jobTitle, employer, salaryInfoAndJobType, URL, fullDesc]
+            writeFile(outdir & jobTitle.replace("/") & ".txt", output)
+            cache.add(URL)
 
-try:
-    salaryGuide = session.findelement("#salaryGuide").get().gettext()
-except UnpackDefect:
-    salaryGuide = "None Listed"
-
-if salaryGuide.contains("Not provided by employer"):
-    salaryGuide = "None Listed"
-
-let fullDesc = session.findElement("#jobDescriptionText").get().getText()
 session.close()
 terminate(chromedriver)
+writeFile(cachefileloc, cache.join("\n"))
 
-#Job Value Scorer
-
-#Parsing Salary
-
-#Output
-var output = """
-\documentclass{article}
-\usepackage[margin=0.7in]{geometry}
-\usepackage{pdfpages}
-\usepackage{hyperref}
-\hypersetup{
-    colorlinks=true,
-    linkcolor=black,
-    filecolor=magenta,      
-    urlcolor=blue,
-    }
-
-
-\begin{document}
-
-\begin{center}
-    \Huge{$1}
-\end{center}
-\hrulefill
-
-    \large{$2} 
-    \hspace{3cm}
-    \large{$3} 
-    \hspace{3cm}
-    \large{\href{$4}{URL}} 
-    \hspace{3cm}
-    \large{$5}
-
-\hrulefill
-
-$6
-
-\end{document}
-""" % [jobTitle, employer, salaryGuide, URL.replace("&","\\&"), jobTimes, fullDesc.replace("#", "\\#").replace("&", "\\&").replace("\"", "\\\"").replace("'","\\'")]
-
-writeFile("/tmp/test.tex", output)
-discard execCmd( "latexmk -pdf -pvc /tmp/test.tex")
-- 
cgit v1.2.3