fix(ohioctapp): parse lower_court

Will help solve freelawproject#1135

- Now parsing lower_court for disambiguating docket numbers across
ohioctapp dockets
- Now parsing per_curiam
- Now parsing parallel_citation for `ohio`
- Updated example files
- Refactored some xpaths

juriscraper/opinions/united_states/state/ohio.py

@@ -13,6 +13,7 @@
 
 from datetime import date
 
+from juriscraper.AbstractSite import logger
 from juriscraper.OpinionSiteLinear import OpinionSiteLinear
 
 
@@ -24,17 +25,12 @@ def __init__(self, *args, **kwargs):
         self.url = "https://www.supremecourtofohio.gov/rod/docs/"
         self.court_id = self.__module__
 
-    def _set_parameters(
-        self,
-        event_validation: str,
-        view_state: str,
-    ) -> None:
+    def _set_parameters(self) -> None:
         """Set the parameters for the search
-
-        :param: event_validation: the event validation token
-        :param: view_state: the view state token
         :return: None
         """
+        event_validation = self.html.xpath("//input[@id='__EVENTVALIDATION']")
+        view_state = self.html.xpath("//input[@id='__VIEWSTATE']")
         self.parameters = {
             "__VIEWSTATEENCRYPTED": "",
             "ctl00$MainContent$ddlCourt": f"{self.court_index}",
@@ -43,43 +39,57 @@ def _set_parameters(
             "ctl00$MainContent$ddlCounty": "0",
             "ctl00$MainContent$btnSubmit": "Submit",
             "ctl00$MainContent$ddlRowsPerPage": "50",
-            "__EVENTVALIDATION": event_validation,
-            "__VIEWSTATE": view_state,
+            "__EVENTVALIDATION": event_validation[0].get("value"),
+            "__VIEWSTATE": view_state[0].get("value"),
         }
-        self.url = "https://www.supremecourt.ohio.gov/rod/docs/"
         self.method = "POST"
 
     def _process_html(self) -> None:
         """Process the HTML and extract the data"""
         if not self.test_mode_enabled():
-            event_validation = self.html.xpath(
-                "//input[@id='__EVENTVALIDATION']"
-            )[0].get("value")
-            view_state = self.html.xpath("//input[@id='__VIEWSTATE']")[0].get(
-                "value"
-            )
-            self._set_parameters(
-                event_validation,
-                view_state,
-            )
+            self._set_parameters()
             self.html = self._download()
 
-        # Skip the header rows and the footer rows
         for row in self.html.xpath(
-            ".//table[@id='MainContent_gvResults']//tr"
-        )[3:-2]:
+            ".//table[@id='MainContent_gvResults']//tr[not(.//a[contains(@href, 'javascript')])]"
+        ):
             # Filter out the case announcements and rulings (ie non-opinions)
-            if not row.xpath(".//td[2]//text()")[0].strip():
+            docket = row.xpath(".//td[2]//text()")[0].strip()
+            name = row.xpath(".//a/text()")[0]
+            if not docket:
+                logger.info("Skipping row with name '%s'", name)
                 continue
-            self.cases.append(
-                {
-                    "judge": row.xpath(".//td[4]//text()")[0],
-                    "docket": row.xpath(".//td[2]//text()")[0],
-                    "date": row.xpath(".//td[6]//text()")[0],
-                    "name": row.xpath(".//a/text()")[0],
-                    "url": row.xpath(".//a")[0].get("href"),
-                    "citation": row.xpath(".//td[8]//text()")[0],
-                    "summary": row.xpath(".//td[3]//text()")[0],
-                    "status": "Published",
-                }
-            )
+
+            judge = row.xpath(".//td[4]//text()")[0]
+            per_curiam = False
+            if "per curiam" in judge.lower():
+                judge = ""
+                per_curiam = True
+
+            citation_or_county = row.xpath(".//td[5]//text()")[0].strip()
+            web_cite = row.xpath(".//td[8]//text()")[0]
+            case = {
+                "docket": docket,
+                "name": name,
+                "judge": judge,
+                "per_curiam": per_curiam,
+                "summary": row.xpath(".//td[3]//text()")[0],
+                "date": row.xpath(".//td[6]//text()")[0],
+                "url": row.xpath(".//a")[0].get("href"),
+                "citation": web_cite,
+                "status": "Published",
+            }
+
+            # Looking for lagged citations like: '175 Ohio St.3d 155'
+            # For Supreme Court cases
+            if self.court_index == 0:
+                citation = ""
+                if web_cite not in citation_or_county:
+                    citation = citation_or_county
+                case["parallel_citation"] = citation
+            elif "ohioctapp" in self.court_id:
+                # For ohioctapp cases. This will be important to disambiguate
+                # docket numbers, which may be repeated across districts
+                case["lower_court"] = f"{citation_or_county} County Court"
+
+            self.cases.append(case)