From 2e1522816022564221ff305d9633ddfe716a93b0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E2=80=9CNaeel=E2=80=9D?= <tazet@narod.ru>
Date: Tue, 23 Jun 2026 12:37:44 +0400
Subject: [PATCH] =?UTF-8?q?v1.0.150:=20revert=20parser.cfm=20to=20working?=
 =?UTF-8?q?=20v1.0.144=20(PDF=E2=86=92VM,=20DOCX=E2=86=92Lucee=20POI)?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 parser.cfm | 52 +++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 39 insertions(+), 13 deletions(-)

diff --git a/parser.cfm b/parser.cfm
index 2607e90..889b250 100644
--- a/parser.cfm
+++ b/parser.cfm
@@ -10,21 +10,47 @@
     <cfargument name="fileBytes" type="binary">
     <cfset var out = {elements: [], errors: []}>
     <cftry>
-        <cfhttp url="https://contracts.kube5s.ru/parse-pdf" method="post" timeout="30">
-            <cfhttpparam type="body" value="#binaryEncode(fileBytes, 'base64')#">
-        </cfhttp>
-        <cfif cfhttp.statusCode EQ 200>
-            <cfset var data = deserializeJSON(cfhttp.fileContent)>
-            <cfif data.ok>
-                <cfset out.elements = data.elements>
-            <cfelse>
-                <cfset arrayAppend(out.errors, "VM parse error: " & (structKeyExists(data, 'error') ? data.error : 'unknown'))>
-            </cfif>
-        <cfelse>
-            <cfset arrayAppend(out.errors, "VM HTTP error: " & cfhttp.statusCode)>
+        <cfset var bais = createObject("java", "java.io.ByteArrayInputStream").init(fileBytes)>
+        <cfset var bis = createObject("java", "java.io.BufferedInputStream").init(bais)>
+        <cfset var pdfDoc = 0>
+        <cftry>
+            <cfset pdfDoc = createObject("java", "org.apache.pdfbox.Loader").loadPDF(bis)>
+            <cfcatch>
+                <cftry>
+                    <cfset pdfDoc = createObject("java", "org.apache.pdfbox.pdmodel.PDDocument").load(bais)>
+                    <cfcatch>
+                        <cfset arrayAppend(out.errors, "PDFBox: cannot open — " & cfcatch.message)>
+                    </cfcatch>
+                </cftry>
+            </cfcatch>
+        </cftry>
+        <cfif NOT isNumeric(pdfDoc)>
+            <cfset var stripper = createObject("java", "org.apache.pdfbox.text.PDFTextStripper")>
+            <cfset var pageCount = pdfDoc.getNumberOfPages()>
+            <cfloop from="1" to="#pageCount#" index="pn">
+                <cfset stripper.setStartPage(pn)>
+                <cfset stripper.setEndPage(pn)>
+                <cfset var pageText = stripper.getText(pdfDoc)>
+                <cfif len(trim(pageText))>
+                    <cfloop list="#pageText#" index="line" delimiters="#chr(10)#">
+                        <cfset line = trim(line)>
+                        <cfif len(line)>
+                            <cfset var el = structNew()>
+                            <cfset el.type = "paragraph">
+                            <cfset el.style = "">
+                            <cfset el.text = line>
+                            <cfset el.page = pn>
+                            <cfset arrayAppend(out.elements, el)>
+                        </cfif>
+                    </cfloop>
+                </cfif>
+            </cfloop>
+            <cfset pdfDoc.close()>
         </cfif>
+        <cfset bis.close()>
+        <cfset bais.close()>
         <cfcatch>
-            <cfset arrayAppend(out.errors, "PDF parse failed: " & cfcatch.message)>
+            <cfset arrayAppend(out.errors, "PDFBox: " & cfcatch.message)>
         </cfcatch>
     </cftry>
     <cfreturn out>