<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">On 07.06.2018 17:35, Ricardo Branco
wrote:<br>
</div>
<blockquote type="cite"
cite="mid:72c80d1b-a662-9383-dc36-4a577eebf57f@wenn.com">
<meta http-equiv="content-type" content="text/html; charset=utf-8">
<p>Solr 7 returns JSON by default but fts_solr requires XML.</p>
<p>Would be good to have wt=xml added to the query to force it to
xml all the time, this would prevent errors if solr has not had
xml set as default for index.<br>
</p>
<p><br>
</p>
<p>Dovecot 2.2.36<br>
Centos 6</p>
<p><b>Doveadm Request:</b></p>
<p>doveadm search -u <a class="moz-txt-link-abbreviated"
href="mailto:caroline@covermg.com" moz-do-not-send="true">caroline@covermg.com</a>
TEXT smith<br>
</p>
<p>doveadm(<a class="moz-txt-link-abbreviated"
href="mailto:caroline@covermg.com" moz-do-not-send="true">caroline@covermg.com</a>):
Error: fts_solr: Invalid XML input at 1:0: not well-formed
(invalid token) (near: {<br>
"responseHeader":{<br>
"status":0,<br>
"QTime":0,<br>
"params":{<br>
"q":"{!lucene q.op=AND}(hdr:smith OR body:smith)",<br>
)<br>
doveadm(<a class="moz-txt-link-abbreviated"
href="mailto:caroline@covermg.com" moz-do-not-send="true">caroline@covermg.com</a>):
Error: fts_solr: Invalid XML input at 1:0: not well-formed
(invalid token) (near: {<br>
"responseHeader":{<br>
"status":0,<br>
"QTime":0,<br>
"params":{<br>
"q":"box:1065b22d6b02185bd40f0000d09efc50 AND user:ca)<br>
<br>
</p>
<p><br>
</p>
<p><b>TCP Dump</b><br>
</p>
<p>Content-Type: application/json;charset=utf-8<br>
Content-Length: 329<br>
<br>
{<br>
"responseHeader":{<br>
"status":0,<br>
"QTime":1,<br>
"params":{<br>
"q":"{!lucene q.op=AND}from:smith",<br>
"fl":"uid,score",<br>
"sort":"uid asc",<br>
"fq":"+box:b571472154d7035b4e250000d09efc50
+user:caroline@covermg.com",<br>
"rows":"69235"}},<br>
"response":{"numFound":0,"start":0,"maxScore":0.0,"docs":[]<br>
}}<br>
<br>
</p>
</blockquote>
<br>
Thanks for reporting this.<br>
<br>
Aki<br>
</body>
</html>